Datasets:

hulx2002
/

VisDocAgentBench

arxiv_id stringlengths 10 10	arxiv_versioned_id stringlengths 12 12	authors listlengths 1 296	categories listlengths 1 5	document_id stringlengths 16 16	license_url stringclasses 4 values	page_count int64 5 139	page_image_access stringclasses 2 values	page_images_included bool 2 classes	pdf_url stringlengths 34 34	primary_category stringclasses 16 values	published timestamp[s]date 2026-01-26 06:38:25 2026-06-26 09:16:55	source_url stringlengths 34 34	title stringlengths 26 122	topic_id stringclasses 10 values	updated timestamp[s]date 2026-01-27 03:18:17 2026-06-26 09:16:55
2601.18203	2601.18203v2	[ "ShunLiang Fu", "Yanxin Zhang", "Yixin Xiang", "Xiaoyu Du", "Jinhui Tang" ]	[ "cs.IR" ]	arxiv_2601_18203	https://creativecommons.org/licenses/by/4.0/	12	included	true	https://arxiv.org/pdf/2601.18203v2	cs.IR	2026-01-26T06:38:25	https://arxiv.org/abs/2601.18203v2	DMAP: Human-Aligned Structural Document Map for Multimodal Document Understanding	t05_multimodal_rag_visual_documents	2026-01-27T03:18:17
2602.06090	2602.06090v1	[ "Xiaoxuan Tang", "Jincheng Wang", "Liwei Luo", "Jingxuan Xu", "Sheng Zhou", "Dajun Chen", "Wei Jiang", "Yong Li" ]	[ "cs.SE", "cs.AI", "cs.CV" ]	arxiv_2602_06090	https://creativecommons.org/licenses/by/4.0/	16	included	true	https://arxiv.org/pdf/2602.06090v1	cs.SE	2026-02-05T06:26:46	https://arxiv.org/abs/2602.06090v1	SVRepair: Structured Visual Reasoning for Automated Program Repair	t03_software_engineering_code_agents	2026-02-05T06:26:46
2602.06593	2602.06593v1	[ "Robert Hutter", "Michael Pradel" ]	[ "cs.SE", "cs.AI" ]	arxiv_2602_06593	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	21	reconstruct_from_source	false	https://arxiv.org/pdf/2602.06593v1	cs.SE	2026-02-06T10:44:09	https://arxiv.org/abs/2602.06593v1	AgentStepper: Interactive Debugging of Software Development Agents	t03_software_engineering_code_agents	2026-02-06T10:44:09
2602.19961	2602.19961v2	[ "Yibo Yan", "Jiahao Huo", "Guanbo Feng", "Mingdong Ou", "Yi Cao", "Xin Zou", "Shuliang Liu", "Yuanhuiyi Lyu", "Yu Huang", "Jungang Li", "Kening Zheng", "Xu Zheng", "Philip S. Yu", "James Kwok", "Xuming Hu" ]	[ "cs.CL", "cs.IR" ]	arxiv_2602_19961	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	35	reconstruct_from_source	false	https://arxiv.org/pdf/2602.19961v2	cs.CL	2026-02-23T15:27:41	https://arxiv.org/abs/2602.19961v2	Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrieval	t05_multimodal_rag_visual_documents	2026-03-23T03:51:40
2602.20502	2602.20502v1	[ "Hongbin Zhong", "Fazle Faisal", "Luis França", "Tanakorn Leesatapornwongsa", "Adriana Szekeres", "Kexin Rong", "Suman Nath" ]	[ "cs.AI", "cs.LG" ]	arxiv_2602_20502	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	12	reconstruct_from_source	false	https://arxiv.org/pdf/2602.20502v1	cs.AI	2026-02-24T03:03:18	https://arxiv.org/abs/2602.20502v1	ActionEngine: From Reactive to Programmatic GUI Agents via State Machine Memory	t02_gui_computer_use_agents	2026-02-24T03:03:18
2602.22013	2602.22013v4	[ "I-Hsiang Chen", "Yu-Wei Liu", "Tse-Yu Wu", "Yu-Chien Chiang", "Jen-Chien Yang", "Wei-Ting Chen" ]	[ "cs.CV" ]	arxiv_2602_22013	https://creativecommons.org/licenses/by/4.0/	10	included	true	https://arxiv.org/pdf/2602.22013v4	cs.CV	2026-02-25T15:27:57	https://arxiv.org/abs/2602.22013v4	RobustVisRAG: Causality-Aware Vision-Based Retrieval-Augmented Generation under Visual Degradations	t05_multimodal_rag_visual_documents	2026-03-26T08:59:16
2602.24134	2602.24134v1	[ "Zhengren Wang", "Dongsheng Ma", "Huaping Zhong", "Jiayu Li", "Wentao Zhang", "Bin Wang", "Conghui He" ]	[ "cs.CV", "cs.CL" ]	arxiv_2602_24134	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	26	reconstruct_from_source	false	https://arxiv.org/pdf/2602.24134v1	cs.CV	2026-02-27T16:09:38	https://arxiv.org/abs/2602.24134v1	AgenticOCR: Parsing Only What You Need for Efficient Retrieval-Augmented Generation	t05_multimodal_rag_visual_documents	2026-02-27T16:09:38
2603.01666	2603.01666v1	[ "Yibo Yan", "Mingdong Ou", "Yi Cao", "Xin Zou", "Shuliang Liu", "Jiahao Huo", "Yu Huang", "James Kwok", "Xuming Hu" ]	[ "cs.CL", "cs.IR" ]	arxiv_2603_01666	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	27	reconstruct_from_source	false	https://arxiv.org/pdf/2603.01666v1	cs.CL	2026-03-02T09:55:00	https://arxiv.org/abs/2603.01666v1	Beyond the Grid: Layout-Informed Multi-Vector Retrieval with Parsed Visual Document Representations	t05_multimodal_rag_visual_documents	2026-03-02T09:55:00
2603.07432	2603.07432v1	[ "Li Gu", "Zihuan Jiang", "Zhixiang Chi", "Huan Liu", "Ziqiang Wang", "Yuanhao Yu", "Glen Berseth", "Yang Wang" ]	[ "cs.CV", "cs.CL", "cs.HC", "cs.LG" ]	arxiv_2603_07432	https://creativecommons.org/licenses/by-nc-nd/4.0/	25	reconstruct_from_source	false	https://arxiv.org/pdf/2603.07432v1	cs.CV	2026-03-08T03:08:11	https://arxiv.org/abs/2603.07432v1	Generalization in Online Reinforcement Learning for Mobile Agents	t02_gui_computer_use_agents	2026-03-08T03:08:11
2603.08013	2603.08013v1	[ "Yuxiang Chai", "Shunye Tang", "Han Xiao", "Rui Liu", "Hongsheng Li" ]	[ "cs.AI" ]	arxiv_2603_08013	https://creativecommons.org/licenses/by/4.0/	16	included	true	https://arxiv.org/pdf/2603.08013v1	cs.AI	2026-03-09T06:41:32	https://arxiv.org/abs/2603.08013v1	PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents	t02_gui_computer_use_agents	2026-03-09T06:41:32
2603.08533	2603.08533v2	[ "Yiping Xie", "Song Chen", "Jingxuan Xing", "Wei Jiang", "Zekun Zhu", "Yingyao Wang", "Pi Bu", "Jun Song", "Yuning Jiang", "Bo Zheng" ]	[ "cs.CV" ]	arxiv_2603_08533	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	16	reconstruct_from_source	false	https://arxiv.org/pdf/2603.08533v2	cs.CV	2026-03-09T16:04:08	https://arxiv.org/abs/2603.08533v2	SecAgent: Efficient Mobile GUI Agent with Semantic Context	t02_gui_computer_use_agents	2026-03-31T03:08:51
2603.12823	2603.12823v1	[ "Xunzhuo Liu", "Bowei He", "Xue Liu", "Andy Luo", "Haichen Zhang", "Huamin Chen" ]	[ "cs.CL", "cs.CV" ]	arxiv_2603_12823	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	16	reconstruct_from_source	false	https://arxiv.org/pdf/2603.12823v1	cs.CL	2026-03-13T09:21:25	https://arxiv.org/abs/2603.12823v1	Adaptive Vision-Language Model Routing for Computer Use Agents	t02_gui_computer_use_agents	2026-03-13T09:21:25
2603.15921	2603.15921v1	[ "Srijan Bansal", "Jiao Fangkai", "Yilun Zhou", "Austin Xu", "Shafiq Joty", "Semih Yavuz" ]	[ "cs.SE", "cs.AI" ]	arxiv_2603_15921	https://creativecommons.org/licenses/by/4.0/	17	included	true	https://arxiv.org/pdf/2603.15921v1	cs.SE	2026-03-16T21:14:28	https://arxiv.org/abs/2603.15921v1	VIBEPASS: Can Vibe Coders Really Pass the Vibe Check?	t03_software_engineering_code_agents	2026-03-16T21:14:28
2603.16085	2603.16085v1	[ "Hui Shan", "Keyang Luo", "Ming Li", "Sizhe Zheng", "Yanwei Fu", "Zhen Chen", "Xiangru Huang" ]	[ "cs.CV", "cs.AI" ]	arxiv_2603_16085	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	25	reconstruct_from_source	false	https://arxiv.org/pdf/2603.16085v1	cs.CV	2026-03-17T03:21:06	https://arxiv.org/abs/2603.16085v1	Interact3D: Compositional 3D Generation of Interactive Objects	t08_3d_spatial_intelligence	2026-03-17T03:21:06
2603.17826	2603.17826v1	[ "Ruize Ma", "Yilei Jiang", "Shilin Zhang", "Zheng Ma", "Yi Feng", "Vincent Ng", "Zhi Wang", "Xiangyu Yue", "Chuanyi Li", "Lewei Lu" ]	[ "cs.SE", "cs.AI" ]	arxiv_2603_17826	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	15	reconstruct_from_source	false	https://arxiv.org/pdf/2603.17826v1	cs.SE	2026-03-18T15:24:22	https://arxiv.org/abs/2603.17826v1	FailureMem: A Failure-Aware Multimodal Framework for Autonomous Software Repair	t03_software_engineering_code_agents	2026-03-18T15:24:22
2603.26211	2603.26211v1	[ "Shrinidhi Kumbhar", "Haofu Liao", "Srikar Appalaraju", "Kunwar Yashraj Singh" ]	[ "cs.CV", "cs.AI" ]	arxiv_2603_26211	https://creativecommons.org/licenses/by/4.0/	14	included	true	https://arxiv.org/pdf/2603.26211v1	cs.CV	2026-03-27T09:34:07	https://arxiv.org/abs/2603.26211v1	Towards GUI Agents: Vision-Language Diffusion Models for GUI Grounding	t02_gui_computer_use_agents	2026-03-27T09:34:07
2604.05012	2604.05012v1	[ "Oteo Mamo", "Olga Kogiou", "Hyunjin Yi", "Weikuan Yu" ]	[ "cs.AR", "cs.AI" ]	arxiv_2604_05012	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	8	reconstruct_from_source	false	https://arxiv.org/pdf/2604.05012v1	cs.AR	2026-04-06T16:00:39	https://arxiv.org/abs/2604.05012v1	Comparative Characterization of KV Cache Management Strategies for LLM Inference	t09_efficient_inference_kv_cache	2026-04-06T16:00:39
2604.14609	2604.14609v1	[ "Zijian Zhang", "Aiwei Yin", "Amaan Baweja", "Jiaru Bai", "Ignacio Gustin", "Varinia Bernales", "Alán Aspuru-Guzik" ]	[ "cs.AI", "physics.comp-ph" ]	arxiv_2604_14609	https://creativecommons.org/licenses/by/4.0/	65	included	true	https://arxiv.org/pdf/2604.14609v1	cs.AI	2026-04-16T04:28:56	https://arxiv.org/abs/2604.14609v1	El Agente Forjador: Task-Driven Agent Generation for Quantum Simulation	t10_scientific_foundation_models	2026-04-16T04:28:56
2604.15821	2604.15821v1	[ "Yuanchang Zhou", "Hongyu Wang", "Yiming Du", "Yan Wang", "Mingzhen Li", "Siyu Hu", "Xiangyu Zhang", "Weijian Liu", "Chen Wang", "Zhuoqiang Guo", "Long Wang", "Jingde Bu", "Yutong Lu", "Guangming Tan", "Weile Jia" ]	[ "cs.DC", "cs.LG" ]	arxiv_2604_15821	https://creativecommons.org/licenses/by/4.0/	11	included	true	https://arxiv.org/pdf/2604.15821v1	cs.DC	2026-04-17T08:20:32	https://arxiv.org/abs/2604.15821v1	Breaking the Training Barrier of Billion-Parameter Universal Machine Learning Interatomic Potentials	t10_scientific_foundation_models	2026-04-17T08:20:32
2604.17288	2604.17288v1	[ "Zizhang Luo", "Yansong Xu", "Runlin Guo", "Fan Cui", "Kexing Zhou", "Mile Xia", "Hongyuan Hou", "Yuhao Luo", "Yun Liang" ]	[ "cs.AR", "cs.AI" ]	arxiv_2604_17288	https://creativecommons.org/licenses/by-nc-nd/4.0/	10	reconstruct_from_source	false	https://arxiv.org/pdf/2604.17288v1	cs.AR	2026-04-19T07:04:49	https://arxiv.org/abs/2604.17288v1	Clover: A Neural-Symbolic Agentic Harness with Stochastic Tree-of-Thoughts for Verified RTL Repair	t03_software_engineering_code_agents	2026-04-19T07:04:49
2604.23276	2604.23276v1	[ "Meizhu Liu", "Yassi Abbasi", "Matthew Rowe", "Michael Avendi", "Paul Li" ]	[ "cs.CV", "cs.AI", "cs.CL" ]	arxiv_2604_23276	https://creativecommons.org/licenses/by/4.0/	8	included	true	https://arxiv.org/pdf/2604.23276v1	cs.CV	2026-04-25T12:40:13	https://arxiv.org/abs/2604.23276v1	Lightweight and Production-Ready PDF Visual Element Parsing	t05_multimodal_rag_visual_documents	2026-04-25T12:40:13
2604.23941	2604.23941v1	[ "Hongxin Li", "Yuntao Chen", "Zhaoxiang Zhang" ]	[ "cs.CV" ]	arxiv_2604_23941	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	20	reconstruct_from_source	false	https://arxiv.org/pdf/2604.23941v1	cs.CV	2026-04-27T01:29:02	https://arxiv.org/abs/2604.23941v1	GoClick: Lightweight Element Grounding Model for Autonomous GUI Interaction	t02_gui_computer_use_agents	2026-04-27T01:29:02
2604.27253	2604.27253v1	[ "Fazle Elahi Faisal", "Qianhui Wu", "Baolin Peng", "Jianfeng Gao" ]	[ "cs.AI" ]	arxiv_2604_27253	https://creativecommons.org/licenses/by/4.0/	21	included	true	https://arxiv.org/pdf/2604.27253v1	cs.AI	2026-04-29T22:57:35	https://arxiv.org/abs/2604.27253v1	AutoSurfer -- Teaching Web Agents through Comprehensive Surfing, Learning, and Modeling	t02_gui_computer_use_agents	2026-04-29T22:57:35
2604.27996	2604.27996v2	[ "Jackson Vonderhorst", "Kuangshi Ai", "Haichao Miao", "Shusen Liu", "Chaoli Wang" ]	[ "cs.AI", "cs.GR", "cs.HC" ]	arxiv_2604_27996	https://creativecommons.org/licenses/by/4.0/	5	included	true	https://arxiv.org/pdf/2604.27996v2	cs.AI	2026-04-30T15:22:28	https://arxiv.org/abs/2604.27996v2	Exploring Interaction Paradigms for LLM Agents in Scientific Visualization	t04_agent_memory_long_horizon	2026-05-13T03:48:48
2605.06460	2605.06460v1	[ "Weien Li", "Rui Song", "Zeyu Li", "Haochen Liu", "Gonghao Zhang", "Difan Jiao", "Zhenwei Tang", "Bowei He", "Haolun Wu", "Xue Liu", "Ye Yuan" ]	[ "cs.LG" ]	arxiv_2605_06460	https://creativecommons.org/licenses/by/4.0/	21	included	true	https://arxiv.org/pdf/2605.06460v1	cs.LG	2026-05-07T15:51:06	https://arxiv.org/abs/2605.06460v1	MINER: Mining Multimodal Internal Representation for Efficient Retrieval	t05_multimodal_rag_visual_documents	2026-05-07T15:51:06
2605.07467	2605.07467v1	[ "Tsuyoshi Okita" ]	[ "cs.LG", "cs.AI", "cs.ET" ]	arxiv_2605_07467	https://creativecommons.org/licenses/by-nc-nd/4.0/	17	reconstruct_from_source	false	https://arxiv.org/pdf/2605.07467v1	cs.LG	2026-05-08T09:14:11	https://arxiv.org/abs/2605.07467v1	Physical Simulators as Do-Operators: Causal Discovery under Latent Confounders for AI-for-Science	t10_scientific_foundation_models	2026-05-08T09:14:11
2605.13986	2605.13986v2	[ "Léo Grinsztajn", "Klemens Flöge", "Oscar Key", "Felix Birkel", "Philipp Jund", "Brendan Roof", "Mihir Manium", "Shi Bin Hoo", "Magnus Bühler", "Anurag Garg", "Dominik Safaric", "Jake Robertson", "Benjamin Jäger", "Simone Alessi", "Adrian Hayler", "Vladyslav Moroshan", "Lennart Puruc...	[ "cs.LG", "stat.ML" ]	arxiv_2605_13986	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	83	reconstruct_from_source	false	https://arxiv.org/pdf/2605.13986v2	cs.LG	2026-05-13T18:01:43	https://arxiv.org/abs/2605.13986v2	TabPFN-3: Technical Report	t06_test_time_scaling_reasoning	2026-05-28T17:22:15
2605.17641	2605.17641v1	[ "Saksham Sahai Srivastava" ]	[ "cs.AI", "cs.CL" ]	arxiv_2605_17641	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	12	reconstruct_from_source	false	https://arxiv.org/pdf/2605.17641v1	cs.AI	2026-05-17T20:21:55	https://arxiv.org/abs/2605.17641v1	Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents	t04_agent_memory_long_horizon	2026-05-17T20:21:55
2605.17965	2605.17965v1	[ "Md Afif Al Mamun", "Gias Uddin" ]	[ "cs.SE", "cs.AI" ]	arxiv_2605_17965	https://creativecommons.org/licenses/by/4.0/	45	included	true	https://arxiv.org/pdf/2605.17965v1	cs.SE	2026-05-18T07:20:13	https://arxiv.org/abs/2605.17965v1	BLAgent: Agentic RAG for File-Level Bug Localization	t03_software_engineering_code_agents	2026-05-18T07:20:13
2605.17976	2605.17976v1	[ "Xinzhe Yuan", "Zhuo Chen", "Jianshu Zhang", "Huan Xiong", "Nanyang Ye", "Yuqiang Li", "Qinying Gu" ]	[ "cs.AI", "math.OC" ]	arxiv_2605_17976	https://creativecommons.org/licenses/by/4.0/	31	included	true	https://arxiv.org/pdf/2605.17976v1	cs.AI	2026-05-18T07:30:25	https://arxiv.org/abs/2605.17976v1	Unleashing LLMs in Bayesian Optimization: Preference-Guided Framework for Scientific Discovery	t10_scientific_foundation_models	2026-05-18T07:30:25
2605.17985	2605.17985v1	[ "Chengjie Hong", "Feixiang He", "Yiheng Zeng", "Lulu Kang", "He Wang" ]	[ "cs.LG", "cs.AI" ]	arxiv_2605_17985	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	22	reconstruct_from_source	false	https://arxiv.org/pdf/2605.17985v1	cs.LG	2026-05-18T07:40:46	https://arxiv.org/abs/2605.17985v1	SAFE-SVD: Sensitivity-Aware Fidelity-Enforcing SVD for Physics Foundation Models	t10_scientific_foundation_models	2026-05-18T07:40:46
2605.19943	2605.19943v1	[ "Amin Sghaier", "Ali Parviz", "Alexia Jolicoeur-Martineau" ]	[ "cs.AI" ]	arxiv_2605_19943	https://creativecommons.org/licenses/by/4.0/	14	included	true	https://arxiv.org/pdf/2605.19943v1	cs.AI	2026-05-19T15:00:38	https://arxiv.org/abs/2605.19943v1	Probabilistic Tiny Recursive Model	t06_test_time_scaling_reasoning	2026-05-19T15:00:38
2605.21312	2605.21312v2	[ "Yicheng Feng", "Xin Tan", "Yangtao Deng", "Yimin Jiang", "Yibo Zhu", "Hong Xu" ]	[ "cs.DC", "cs.AI", "cs.LG" ]	arxiv_2605_21312	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	20	reconstruct_from_source	false	https://arxiv.org/pdf/2605.21312v2	cs.DC	2026-05-20T15:40:18	https://arxiv.org/abs/2605.21312v2	Frontier: Towards Comprehensive and Accurate LLM Inference Simulation	t09_efficient_inference_kv_cache	2026-06-13T06:07:21
2605.22829	2605.22829v1	[ "Yifan Zhu", "Yu Mi", "Yue Lu", "Yanchu Guan", "Zhixuan Chu" ]	[ "cs.IR", "cs.AI" ]	arxiv_2605_22829	https://creativecommons.org/licenses/by/4.0/	19	included	true	https://arxiv.org/pdf/2605.22829v1	cs.IR	2026-04-18T05:04:49	https://arxiv.org/abs/2605.22829v1	LFRAG: Layout-oriented Fine-grained Retrieval-Augmented Generation on Multimodal Document Understanding	t05_multimodal_rag_visual_documents	2026-04-18T05:04:49
2605.23057	2605.23057v1	[ "Aman Sunesh", "Ali Alshehhi", "Hivansh Dhakne" ]	[ "cs.LG", "cs.CL", "cs.PF" ]	arxiv_2605_23057	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	11	reconstruct_from_source	false	https://arxiv.org/pdf/2605.23057v1	cs.LG	2026-05-21T21:46:57	https://arxiv.org/abs/2605.23057v1	ModeSwitch-LLM: A Lightweight Phase-Aware Controller for Cross-Mode LLM Inference on a Single GPU	t09_efficient_inference_kv_cache	2026-05-21T21:46:57
2605.23163	2605.23163v2	[ "Kewei Zhang", "Jin Wang", "Sensen Gao", "Chengyue Wu", "Yulong Cao", "Songyang Han", "Boris Ivanovic", "Langechuan Liu", "Marco Pavone", "Song Han", "Daquan Zhou", "Enze Xie" ]	[ "cs.CL" ]	arxiv_2605_23163	https://creativecommons.org/licenses/by-nc-nd/4.0/	17	reconstruct_from_source	false	https://arxiv.org/pdf/2605.23163v2	cs.CL	2026-05-22T02:31:32	https://arxiv.org/abs/2605.23163v2	Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving	t09_efficient_inference_kv_cache	2026-05-25T07:32:46
2605.27570	2605.27570v1	[ "Gabriele Cesa", "Thomas Hehn", "Aleix Torres-Camps", "Àlex Batlle Casellas", "Jordi Ros-Giralt", "Arash Behboodi", "Tribhuvanesh Orekondy" ]	[ "cs.AI" ]	arxiv_2605_27570	https://creativecommons.org/licenses/by/4.0/	22	included	true	https://arxiv.org/pdf/2605.27570v1	cs.AI	2026-05-26T18:43:15	https://arxiv.org/abs/2605.27570v1	LaneRoPE: Positional Encoding for Collaborative Parallel Reasoning and Generation	t06_test_time_scaling_reasoning	2026-05-26T18:43:15
2605.29074	2605.29074v1	[ "Jiyao Zhang", "Mingxu Zhang", "Yitong Peng", "Haoxuan Liu", "Chenshuo Wang", "Yuxing Long", "Haoyang Huang", "Dongjiang Li", "Nan Duan", "Hui Shen", "Hao Dong" ]	[ "cs.CV", "cs.RO" ]	arxiv_2605_29074	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	11	reconstruct_from_source	false	https://arxiv.org/pdf/2605.29074v1	cs.CV	2026-05-27T20:28:56	https://arxiv.org/abs/2605.29074v1	Embodied3DBench: Benchmarking Low-Level Embodied Spatial Intelligence of Vision Language Models	t08_3d_spatial_intelligence	2026-05-27T20:28:56
2605.29639	2605.29639v1	[ "Boyu Tan", "Jiarui Guo", "Zongwei Lv", "Hanbo Sun", "Tong Yang", "Kan Liu", "Xinfei Shi", "Zetao Hu", "Yaxin Yu", "Chi Zhang", "Jianning Zhang", "Xi Yang", "Wei Zhang", "Bo Cai", "Silu Zhou", "Xiyu Wang", "Na He", "Yinghao Yu", "Wending Bao", "Guiyang Huang", "Yuxing Yuan", ...	[ "cs.OS" ]	arxiv_2605_29639	https://creativecommons.org/licenses/by/4.0/	14	included	true	https://arxiv.org/pdf/2605.29639v1	cs.OS	2026-05-28T09:07:06	https://arxiv.org/abs/2605.29639v1	RTP-LLM: High-Performance Alibaba LLM Inference Engine	t09_efficient_inference_kv_cache	2026-05-28T09:07:06
2605.30917	2605.30917v1	[ "Gyu-Hwung Cho", "Youngjune Lee", "Kiyoon Jeong", "Siyoung Lee", "Sanggyu Han", "Hervé Dejean", "Stéphane Clinchant", "Seung-won Hwang" ]	[ "cs.IR", "cs.CV" ]	arxiv_2605_30917	https://creativecommons.org/licenses/by-nc-sa/4.0/	12	included	true	https://arxiv.org/pdf/2605.30917v1	cs.IR	2026-05-29T07:01:45	https://arxiv.org/abs/2605.30917v1	Inference-Free Multimodal Learned Sparse Retrieval for Production-Scale Visual Document Search	t05_multimodal_rag_visual_documents	2026-05-29T07:01:45
2606.00756	2606.00756v1	[ "Yannan Wang", "Longli Yang", "Zhen Liu", "Abhishek Kumar", "Carsten Maple" ]	[ "cs.AI" ]	arxiv_2606_00756	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	19	reconstruct_from_source	false	https://arxiv.org/pdf/2606.00756v1	cs.AI	2026-05-30T14:45:39	https://arxiv.org/abs/2606.00756v1	CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems	t04_agent_memory_long_horizon	2026-05-30T14:45:39
2606.00832	2606.00832v1	[ "Adril Putra Merin", "David Anugraha", "Ayu Purwarianti", "Genta Indra Winata" ]	[ "cs.CL" ]	arxiv_2606_00832	https://creativecommons.org/licenses/by/4.0/	9	included	true	https://arxiv.org/pdf/2606.00832v1	cs.CL	2026-05-30T18:08:51	https://arxiv.org/abs/2606.00832v1	Momento: Evaluating Persistent Memory and Reasoning with Multi-Session Agentic Conversations	t04_agent_memory_long_horizon	2026-05-30T18:08:51
2606.01613	2606.01613v2	[ "Kanwar Bharat Singh" ]	[ "cs.IR", "cs.AI", "cs.MA" ]	arxiv_2606_01613	https://creativecommons.org/licenses/by/4.0/	24	included	true	https://arxiv.org/pdf/2606.01613v2	cs.IR	2026-06-01T03:01:51	https://arxiv.org/abs/2606.01613v2	TechRAG: Evidence-Gated Multimodal Agentic RAG for Technical Literature Reasoning	t05_multimodal_rag_visual_documents	2026-06-13T19:11:55
2606.02355	2606.02355v1	[ "Zhongyu He", "Yuanfan Li", "Fei Huang", "Tianyu Chen", "Siyuan Chen", "Xingyang Li", "Meng Hsuan Yu", "Xiangrong Liu", "Leyi Wei", "Lu Pan", "Ke Zeng", "Xunliang Cai" ]	[ "cs.AI", "cs.LG" ]	arxiv_2606_02355	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	15	reconstruct_from_source	false	https://arxiv.org/pdf/2606.02355v1	cs.AI	2026-06-01T15:02:59	https://arxiv.org/abs/2606.02355v1	SIRI: Self-Internalizing Reinforcement Learning with Intrinsic Skills for LLM Agent Training	t04_agent_memory_long_horizon	2026-06-01T15:02:59
2606.02800	2606.02800v4	[ "NVIDIA", ":", "Aditi", "Niket Agarwal", "Arslan Ali", "Jon Allen", "Martin Antolini", "Adeline Aubame", "Alisson Azzolini", "Junjie Bai", "Maciej Bala", "Yogesh Balaji", "Josh Bapst", "Aarti Basant", "Mukesh Beladiya", "Mohammad Qazim Bhat", "Zaid Pervaiz Bhat", "Dan Blick", "Va...	[ "cs.CV", "cs.AI", "cs.LG", "cs.MM", "cs.RO" ]	arxiv_2606_02800	https://creativecommons.org/licenses/by/4.0/	139	included	true	https://arxiv.org/pdf/2606.02800v4	cs.CV	2026-06-01T19:12:30	https://arxiv.org/abs/2606.02800v4	Cosmos 3: Omnimodal World Models for Physical AI	t01_embodied_world_models	2026-06-23T17:33:32
2606.03099	2606.03099v1	[ "Kailin Lyu", "Zhiqiang Yuan", "Jianwei He", "Qiwei Yan", "Xuanbo Su", "Nanxing Hu", "Yang Liu", "Ce Hao", "Shengqian Qin", "Lianyu Hu", "Jinchao Zhang", "Jie Zhou" ]	[ "cs.CL", "cs.AI" ]	arxiv_2606_03099	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	17	reconstruct_from_source	false	https://arxiv.org/pdf/2606.03099v1	cs.CL	2026-06-02T03:38:44	https://arxiv.org/abs/2606.03099v1	PhotoCraft: Agentic Reasoning with Hierarchical Self-Evolving Memory for Deep Image Search	t04_agent_memory_long_horizon	2026-06-02T03:38:44
2606.04432	2606.04432v1	[ "Thanh-Tung Le", "Yunhan Zhao", "Menglei Chai", "Zhengyang Shen", "Zhe Cao", "Danhang Tang", "Xiaohui Xie", "Deying Kong" ]	[ "cs.CV" ]	arxiv_2606_04432	https://creativecommons.org/licenses/by-nc-sa/4.0/	13	included	true	https://arxiv.org/pdf/2606.04432v1	cs.CV	2026-06-03T04:25:11	https://arxiv.org/abs/2606.04432v1	DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation	t07_video_generation_control	2026-06-03T04:25:11
2606.04527	2606.04527v1	[ "Yuxuan Bian", "Zeyue Xue", "Songchun Zhang", "Shiyi Zhang", "Weiyang Jin", "Yaowei Li", "Junhao Zhuang", "Haoran Li", "Jie Huang", "Haoyang Huang", "Nan Duan", "Qiang Xu" ]	[ "cs.MM", "cs.CV", "cs.GR" ]	arxiv_2606_04527	https://creativecommons.org/licenses/by/4.0/	21	included	true	https://arxiv.org/pdf/2606.04527v1	cs.MM	2026-06-03T07:09:01	https://arxiv.org/abs/2606.04527v1	Echo-Infinity: Learning Evolving Memory for Real-Time Infinite Video Generation	t07_video_generation_control	2026-06-03T07:09:01
2606.05145	2606.05145v1	[ "Nizar Islah", "Istabrak Abbes", "Irina Rish", "Sarath Chandar", "Eilif B. Muller" ]	[ "cs.LG", "cs.AI", "cs.CL" ]	arxiv_2606_05145	https://creativecommons.org/licenses/by/4.0/	43	included	true	https://arxiv.org/pdf/2606.05145v1	cs.LG	2026-06-03T17:50:26	https://arxiv.org/abs/2606.05145v1	Failed Reasoning Traces Tell You What Is Fixable (But Not by Reading Them)	t06_test_time_scaling_reasoning	2026-06-03T17:50:26
2606.05761	2606.05761v2	[ "Wenxuan Wang", "Haoyu Sun", "Fukuan Hou", "Mingyang Song", "Weinan Zhang", "Yu Cheng", "Yang Yang" ]	[ "cs.AI", "cs.CL" ]	arxiv_2606_05761	https://creativecommons.org/licenses/by/4.0/	47	included	true	https://arxiv.org/pdf/2606.05761v2	cs.AI	2026-06-04T06:43:11	https://arxiv.org/abs/2606.05761v2	SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents	t04_agent_memory_long_horizon	2026-06-05T04:28:58
2606.05773	2606.05773v1	[ "Chong Ma", "Taiyi Su", "Jian Zhu", "Jianjun Zhang", "Zitai Huang", "Yi Xu", "Hanli Wang" ]	[ "cs.RO" ]	arxiv_2606_05773	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	14	reconstruct_from_source	false	https://arxiv.org/pdf/2606.05773v1	cs.RO	2026-06-04T06:57:25	https://arxiv.org/abs/2606.05773v1	PiL-World: A Chunk-Wise World Model for VLA Policy-in-the-Loop Evaluation	t01_embodied_world_models	2026-06-04T06:57:25
2606.06322	2606.06322v1	[ "Nathan Bout", "Maxime Langevin", "Ronan Riochet" ]	[ "cs.AI" ]	arxiv_2606_06322	https://creativecommons.org/licenses/by/4.0/	19	included	true	https://arxiv.org/pdf/2606.06322v1	cs.AI	2026-06-04T15:57:29	https://arxiv.org/abs/2606.06322v1	DragOn: A Benchmark and Dataset for Drag-Based GUI Interactions	t02_gui_computer_use_agents	2026-06-04T15:57:29
2606.06891	2606.06891v1	[ "Hanxun Yu", "Xuan Qu", "Lei Ke", "Boqiang Zhang", "Yuxin Wang", "Jianke Zhu", "Dong Yu" ]	[ "cs.CV" ]	arxiv_2606_06891	https://creativecommons.org/licenses/by/4.0/	32	included	true	https://arxiv.org/pdf/2606.06891v1	cs.CV	2026-06-05T04:16:24	https://arxiv.org/abs/2606.06891v1	Stream3D-VLM: Online 3D Spatial Understanding with Incremental Geometry Priors	t08_3d_spatial_intelligence	2026-06-05T04:16:24
2606.06915	2606.06915v2	[ "Vladislav Smirnov", "Chieu Nguyen", "Sergey Senichev", "Minh Ngoc Ta", "Ekaterina Fadeeva", "Artem Vazhentsev", "Daria Galimzianova", "Nikolai Rozanov", "Viktor Mazanov", "Jingwei Ni", "Tianyi Wu", "Igor Kiselev", "Mrinmaya Sachan", "Iryna Gurevych", "Preslav Nakov", "Timothy Baldwin"...	[ "cs.CL", "cs.AI", "cs.LG" ]	arxiv_2606_06915	https://creativecommons.org/licenses/by/4.0/	15	included	true	https://arxiv.org/pdf/2606.06915v2	cs.CL	2026-06-05T05:28:46	https://arxiv.org/abs/2606.06915v2	ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning	t06_test_time_scaling_reasoning	2026-06-08T11:00:40
2606.08688	2606.08688v1	[ "Chunji Lv", "Jiaxi Ye", "Yuchen Jiang", "Rexar Lin", "Changsheng Li" ]	[ "cs.RO", "cs.CV" ]	arxiv_2606_08688	https://creativecommons.org/licenses/by/4.0/	20	included	true	https://arxiv.org/pdf/2606.08688v1	cs.RO	2026-06-07T15:42:56	https://arxiv.org/abs/2606.08688v1	PhysAgent: Automating Physics-Based 4D Synthesis via Trajectory-Grounded Multi-Agent Feedback	t10_scientific_foundation_models	2026-06-07T15:42:56
2606.08737	2606.08737v1	[ "Yunfan Lou", "Yifan Ye", "Yankai Fu", "Jun Cen", "Xiaowei Chi", "Yaoxu Lyu", "Peidong Jia", "Sirui Han", "Zhihe Lu", "Shanghang Zhang" ]	[ "cs.RO" ]	arxiv_2606_08737	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	16	reconstruct_from_source	false	https://arxiv.org/pdf/2606.08737v1	cs.RO	2026-06-07T17:18:23	https://arxiv.org/abs/2606.08737v1	Dream-Tac: A Unified Tactile World Action Model for Contact-Rich Robot Manipulation	t01_embodied_world_models	2026-06-07T17:18:23
2606.09864	2606.09864v1	[ "Bruce Changlong Xu", "Adarsh Kumarappan", "Mu Zhou" ]	[ "cs.LG", "cs.AI", "cs.ET" ]	arxiv_2606_09864	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	61	reconstruct_from_source	false	https://arxiv.org/pdf/2606.09864v1	cs.LG	2026-06-01T02:02:20	https://arxiv.org/abs/2606.09864v1	Alignment Collapse Under KV Cache Quantization: Diagnosis and Mitigation	t09_efficient_inference_kv_cache	2026-06-01T02:02:20
2606.10183	2606.10183v1	[ "Konstantin Kuklev", "Viacheslav Vasilev", "Alexander Kunitsyn", "Andrei Ivaniuta", "Denis Dimitrov" ]	[ "cs.CV", "cs.AI", "cs.MM" ]	arxiv_2606_10183	https://creativecommons.org/licenses/by/4.0/	10	included	true	https://arxiv.org/pdf/2606.10183v1	cs.CV	2026-06-08T21:21:01	https://arxiv.org/abs/2606.10183v1	Making Time Editable in Video Diffusion Transformers	t07_video_generation_control	2026-06-08T21:21:01
2606.10662	2606.10662v1	[ "Yuzhen Mao", "Azalia Mirhoseini" ]	[ "cs.MA", "cs.AI" ]	arxiv_2606_10662	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	23	reconstruct_from_source	false	https://arxiv.org/pdf/2606.10662v1	cs.MA	2026-06-09T10:13:07	https://arxiv.org/abs/2606.10662v1	Decentralized Multi-Agent Systems with Shared Context	t06_test_time_scaling_reasoning	2026-06-09T10:13:07
2606.11078	2606.11078v1	[ "Jaewoo Lee", "Zaid Khan", "Archiki Prasad", "Justin Chih-Yao Chen", "Supriyo Chakraborty", "Kartik Balasubramaniam", "Sambit Sahu", "Elias Stengel-Eskin", "Hyunji Lee", "Mohit Bansal" ]	[ "cs.AI", "cs.CL", "cs.CV" ]	arxiv_2606_11078	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	29	reconstruct_from_source	false	https://arxiv.org/pdf/2606.11078v1	cs.AI	2026-06-09T16:39:10	https://arxiv.org/abs/2606.11078v1	A History-Aware Visually Grounded Critic for Computer Use Agents	t02_gui_computer_use_agents	2026-06-09T16:39:10
2606.11576	2606.11576v1	[ "Ahmadreza Jeddi", "Minh Ngoc Le", "Amirhossein Kazerouni", "Hakki Can Karaimer", "Hue Nguyen", "Iqbal Mohomed", "Michael Brudno", "Alex Levinshtein", "Konstantinos G. Derpanis", "Babak Taati", "Radek Grzeszczuk" ]	[ "cs.CV", "cs.AI" ]	arxiv_2606_11576	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	30	reconstruct_from_source	false	https://arxiv.org/pdf/2606.11576v1	cs.CV	2026-06-10T02:06:47	https://arxiv.org/abs/2606.11576v1	AVIS: Adaptive Test-Time Scaling for Vision-Language Models	t06_test_time_scaling_reasoning	2026-06-10T02:06:47
2606.13035	2606.13035v1	[ "Yu Meng", "Xiangyang Luo", "Letian Li", "Wenyuan Jiang", "Chen Gao", "Xinlei Chen", "Yong Li", "Xiao-Ping Zhang" ]	[ "cs.CV", "cs.AI" ]	arxiv_2606_13035	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	17	reconstruct_from_source	false	https://arxiv.org/pdf/2606.13035v1	cs.CV	2026-06-11T08:16:08	https://arxiv.org/abs/2606.13035v1	TetherCache: Stabilizing Autoregressive Long-Form Video Generation with Gated Recall and Trusted Alignment	t07_video_generation_control	2026-06-11T08:16:08
2606.13494	2606.13494v1	[ "Daichi Azuma", "Taiki Miyanishi", "Koya Sakamoto", "Shuhei Kurita", "Yaonan Zhu", "Petr Khrapchenkov", "Motoaki Kawanabe", "Yusuke Iwasawa", "Yutaka Matsuo" ]	[ "cs.RO", "cs.CV" ]	arxiv_2606_13494	https://creativecommons.org/licenses/by/4.0/	22	included	true	https://arxiv.org/pdf/2606.13494v1	cs.RO	2026-06-11T15:44:36	https://arxiv.org/abs/2606.13494v1	NavWAM: A Navigation World Action Model for Goal-Conditioned Visual Navigation	t01_embodied_world_models	2026-06-11T15:44:36
2606.13672	2606.13672v2	[ "Arnav Kumar Jain", "Yilin Wu", "Jesse Farebrother", "Gokul Swamy", "Andrea Bajcsy" ]	[ "cs.RO" ]	arxiv_2606_13672	https://creativecommons.org/licenses/by/4.0/	32	included	true	https://arxiv.org/pdf/2606.13672v2	cs.RO	2026-06-11T17:59:15	https://arxiv.org/abs/2606.13672v2	WEAVER, Better, Faster, Longer: An Effective World Model for Robotic Manipulation	t01_embodied_world_models	2026-06-16T20:54:22
2606.13740	2606.13740v1	[ "Tuowei Wang", "Yanfan Sun", "Ju Ren" ]	[ "cs.LG" ]	arxiv_2606_13740	https://creativecommons.org/licenses/by/4.0/	14	included	true	https://arxiv.org/pdf/2606.13740v1	cs.LG	2026-06-11T12:44:57	https://arxiv.org/abs/2606.13740v1	Efficient On-Device Diffusion LLM Inference with Mobile NPU	t09_efficient_inference_kv_cache	2026-06-11T12:44:57
2606.13768	2606.13768v2	[ "Sharath Girish", "Tsai-Shien Chen", "Zhikang Dong", "Mukesh Singhal", "Hao Chen", "Sergey Tulyakov", "Aliaksandr Siarohin" ]	[ "cs.CV", "cs.AI" ]	arxiv_2606_13768	https://creativecommons.org/licenses/by/4.0/	32	included	true	https://arxiv.org/pdf/2606.13768v2	cs.CV	2026-06-11T17:59:10	https://arxiv.org/abs/2606.13768v2	CineOrchestra: Unified Entity-Centric Conditioning for Cinematic Video Generation	t07_video_generation_control	2026-06-16T17:59:39
2606.14162	2606.14162v1	[ "Xunzhi Xiang", "Zixuan Duan", "Yabo Chen", "Zhengxuan Wei", "Guiyu Zhang", "Zixiao Gu", "Zhe Gao", "Haibin Huang", "Chi Zhang", "Qi Fan", "Xuelong Li" ]	[ "cs.CV" ]	arxiv_2606_14162	https://creativecommons.org/licenses/by/4.0/	17	included	true	https://arxiv.org/pdf/2606.14162v1	cs.CV	2026-06-12T06:41:13	https://arxiv.org/abs/2606.14162v1	VideoWeave: Unlocking Geometric Consistency in Video Generation via Joint Geometry-Video Modeling	t07_video_generation_control	2026-06-12T06:41:13
2606.14870	2606.14870v1	[ "Ibrahim Elsharkawy", "Joschka Birk", "Vinicius Mikuni", "Wahid Bhimji", "Gregor Kasieczka", "Benjamin Nachman" ]	[ "hep-ph", "cs.LG" ]	arxiv_2606_14870	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	33	reconstruct_from_source	false	https://arxiv.org/pdf/2606.14870v1	hep-ph	2026-06-12T18:15:04	https://arxiv.org/abs/2606.14870v1	Pre-Training for Simulation-Based Science: A Study on Jet Foundation Model Training Objectives	t10_scientific_foundation_models	2026-06-12T18:15:04
2606.14948	2606.14948v1	[ "Kirill Vasilevski", "Ximing Dong", "Benjamin Rombaut", "Ruochen Deng", "Jiahuei Lin", "Arthur Leung", "Dayi Lin", "Boyuan Chen", "Shaowei Wang", "Ahmed E. Hassan" ]	[ "cs.SE", "cs.AI" ]	arxiv_2606_14948	https://creativecommons.org/licenses/by-nc-nd/4.0/	19	reconstruct_from_source	false	https://arxiv.org/pdf/2606.14948v1	cs.SE	2026-06-12T20:46:04	https://arxiv.org/abs/2606.14948v1	Beyond Correctness: Enhancing Architectural Reasoning in Code LLMs via Scalable Labeling with Agentic Judgment	t03_software_engineering_code_agents	2026-06-12T20:46:04
2606.15001	2606.15001v1	[ "Xiaoyu Wang", "Bingqing Cheng" ]	[ "physics.comp-ph", "cond-mat.mtrl-sci", "cs.LG", "physics.chem-ph" ]	arxiv_2606_15001	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	15	reconstruct_from_source	false	https://arxiv.org/pdf/2606.15001v1	physics.comp-ph	2026-06-12T22:45:20	https://arxiv.org/abs/2606.15001v1	Distilling latent electrostatics from foundation machine learning interatomic potentials	t10_scientific_foundation_models	2026-06-12T22:45:20
2606.15032	2606.15032v1	[ "Yang Yu", "Shiyuan Zhang", "Yifei Sheng", "Haoxiang Ren", "Haoxin Lin" ]	[ "cs.LG" ]	arxiv_2606_15032	https://creativecommons.org/licenses/by/4.0/	24	included	true	https://arxiv.org/pdf/2606.15032v1	cs.LG	2026-06-13T00:21:21	https://arxiv.org/abs/2606.15032v1	How Should World Models Be Evaluated? A Decision-Making-Centric Position	t01_embodied_world_models	2026-06-13T00:21:21
2606.17257	2606.17257v1	[ "Rohit Kundu", "Arindam Dutta", "Sarosij Bose", "Athula Balachandran", "Amit K. Roy-Chowdhury" ]	[ "cs.CV", "cs.AI" ]	arxiv_2606_17257	https://creativecommons.org/licenses/by/4.0/	17	included	true	https://arxiv.org/pdf/2606.17257v1	cs.CV	2026-06-15T20:03:09	https://arxiv.org/abs/2606.17257v1	Pulling The REINS: Training-Free Safety Alignment of Video Diffusion Models via Representation Steering	t07_video_generation_control	2026-06-15T20:03:09
2606.17803	2606.17803v1	[ "Vaggelis Dorovatas", "Nancy Kalaj", "Rahaf Aljundi" ]	[ "cs.LG" ]	arxiv_2606_17803	https://creativecommons.org/licenses/by/4.0/	17	included	true	https://arxiv.org/pdf/2606.17803v1	cs.LG	2026-06-16T11:27:28	https://arxiv.org/abs/2606.17803v1	Continual Self-Improvement with Lightweight Experiential Latent Memories	t06_test_time_scaling_reasoning	2026-06-16T11:27:28
2606.18180	2606.18180v1	[ "Rongjin Guo", "Dong Liang", "Yuhao Liu", "Fang Liu", "Tianyu Huang", "Gerhard P. Hancke", "Rynson W. H. Lau" ]	[ "cs.CV" ]	arxiv_2606_18180	https://creativecommons.org/licenses/by/4.0/	16	included	true	https://arxiv.org/pdf/2606.18180v1	cs.CV	2026-06-16T17:13:58	https://arxiv.org/abs/2606.18180v1	EgoCS-400K: An Egocentric Gameplay Dataset for World Models	t01_embodied_world_models	2026-06-16T17:13:58
2606.18375	2606.18375v3	[ "Yuhang Huang", "Xuan Lv", "Junyan Xu", "Zhiyuan Yu", "Jiazhao Zhang", "Ruizhen Hu", "Wancheng Feng", "Shilong Zou", "Hewen Xiao", "Ziqiao Zhou", "Kaiyun Huang", "Zhiyu Peng", "Juzhan Xu", "Hang Zhao", "Chenyang Zhu", "Renjiao Yi", "Yifei Huang", "Douhui Wu", "Yan Zhang", "Kexu...	[ "cs.RO" ]	arxiv_2606_18375	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	20	reconstruct_from_source	false	https://arxiv.org/pdf/2606.18375v3	cs.RO	2026-06-16T18:23:23	https://arxiv.org/abs/2606.18375v3	PAIWorld: A 3D-Consistent World Foundation Model for Robotic Manipulation	t01_embodied_world_models	2026-06-23T16:50:09
2606.18478	2606.18478v2	[ "Siyi Chen", "Shaowei Liu", "Yixuan Jia", "Zian Wang", "Huan Ling", "Qing Qu", "Jun Gao" ]	[ "cs.CV" ]	arxiv_2606_18478	https://creativecommons.org/licenses/by/4.0/	34	included	true	https://arxiv.org/pdf/2606.18478v2	cs.CV	2026-06-16T20:38:30	https://arxiv.org/abs/2606.18478v2	Data-Forcing Distillation: Restoring Diversity and Fidelity in Few-Step Video Generation	t07_video_generation_control	2026-06-23T08:04:55
2606.18847	2606.18847v1	[ "Yehang Zhang", "Jianchong Su", "Haojian Huang", "Yifan Chang", "Tianhao Zhou", "Xinli Xu", "Yingjie Xu", "Yinchuan Li", "Zexi Li", "Ying-Cong Chen" ]	[ "cs.AI" ]	arxiv_2606_18847	https://creativecommons.org/licenses/by/4.0/	27	included	true	https://arxiv.org/pdf/2606.18847v1	cs.AI	2026-06-17T09:26:26	https://arxiv.org/abs/2606.18847v1	WorldLines: Benchmarking and Modeling Long-Horizon Stateful Embodied Agents	t04_agent_memory_long_horizon	2026-06-17T09:26:26
2606.19340	2606.19340v2	[ "Jisoo Kim", "Sangwon Baik", "Taeksoo Kim", "Sungjoo Kim", "Junyoung Lee", "Mingi Choi", "Hanbyul Joo" ]	[ "cs.RO" ]	arxiv_2606_19340	https://creativecommons.org/licenses/by/4.0/	21	included	true	https://arxiv.org/pdf/2606.19340v2	cs.RO	2026-06-17T17:59:56	https://arxiv.org/abs/2606.19340v2	ZeroDex: Zero-Shot Long-Horizon Dexterous Manipulation via Multi-View 3D-Grounded VLM Reasoning	t08_3d_spatial_intelligence	2026-06-19T13:42:23
2606.19531	2606.19531v1	[ "Yuyang Zhang", "Wenyao Zhang", "Zekun Qi", "He Zhang", "Haitao Lin", "Jingbo Zhang", "Yao Mu", "Xiaokang Yang", "Wenjun Zeng", "Xin Jin" ]	[ "cs.CV", "cs.RO" ]	arxiv_2606_19531	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	19	reconstruct_from_source	false	https://arxiv.org/pdf/2606.19531v1	cs.CV	2026-06-17T19:25:28	https://arxiv.org/abs/2606.19531v1	ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?	t01_embodied_world_models	2026-06-17T19:25:28
2606.19746	2606.19746v1	[ "Ruiyang Ma", "Teng Ma", "Junru Li", "Hantian Zha", "Xuchun Shang", "Qingda Hu", "Zheng Liu", "Xinjun Yang", "Tao Ma", "Guojie Luo" ]	[ "cs.DC" ]	arxiv_2606_19746	https://creativecommons.org/licenses/by/4.0/	22	included	true	https://arxiv.org/pdf/2606.19746v1	cs.DC	2026-06-18T03:21:32	https://arxiv.org/abs/2606.19746v1	SAC: Disaggregated KV Cache System for Sparse Attention LLMs with CXL	t09_efficient_inference_kv_cache	2026-06-18T03:21:32
2606.20512	2606.20512v2	[ "Asa Shepard", "Jeannie Albrecht" ]	[ "cs.SE", "cs.LG" ]	arxiv_2606_20512	https://creativecommons.org/licenses/by/4.0/	20	included	true	https://arxiv.org/pdf/2606.20512v2	cs.SE	2026-06-18T17:30:15	https://arxiv.org/abs/2606.20512v2	Probe-and-Refine Tuning of Repository Guidance for Coding Agents	t03_software_engineering_code_agents	2026-06-19T10:59:07
2606.20729	2606.20729v1	[ "Masaya Hagai", "Yuta Suzuki", "Tomoya Murata", "Shuhei Kurita", "Masaki Adachi" ]	[ "physics.chem-ph", "cond-mat.mtrl-sci", "cs.AI", "cs.LG" ]	arxiv_2606_20729	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	17	reconstruct_from_source	false	https://arxiv.org/pdf/2606.20729v1	physics.chem-ph	2026-06-17T05:58:58	https://arxiv.org/abs/2606.20729v1	LLM-Guided Test-Time Discovery of Quantum-Chemical Approximation Algorithms	t10_scientific_foundation_models	2026-06-17T05:58:58
2606.20891	2606.20891v1	[ "Koichi Namekata", "Yash Kant", "Zhizheng Liu", "Ryan D Burgert", "Yuancheng Xu", "Kuan Heng Lin", "Emmett Steven", "Julien Philip", "Li Ma", "Andrea Vedaldi", "Paul Debevec", "Ning Yu" ]	[ "cs.CV", "cs.LG" ]	arxiv_2606_20891	https://creativecommons.org/licenses/by/4.0/	29	included	true	https://arxiv.org/pdf/2606.20891v1	cs.CV	2026-06-18T19:40:03	https://arxiv.org/abs/2606.20891v1	Go-with-the-Track: Video Compositing and Motion Control with Point Tracking	t07_video_generation_control	2026-06-18T19:40:03
2606.21023	2606.21023v1	[ "Zhenting Zhu", "Lucas Thai", "Shan Yu", "Yicheng Liu", "Yifan Qiao", "Chenxi Wang", "Harry Xu", "Junyi Shu" ]	[ "cs.LG" ]	arxiv_2606_21023	https://creativecommons.org/licenses/by/4.0/	25	included	true	https://arxiv.org/pdf/2606.21023v1	cs.LG	2026-06-19T01:21:10	https://arxiv.org/abs/2606.21023v1	Demystifying Numerical Instability in LLM Inference: Achieving Reproducible Inference for Mission-Critical Tasks with HEAL	t09_efficient_inference_kv_cache	2026-06-19T01:21:10
2606.21633	2606.21633v1	[ "Omin Kwon", "Doyeon Kim", "Jongseok Park", "Seung Yul Lee", "Ion Stoica", "Jae W. Lee" ]	[ "cs.LG" ]	arxiv_2606_21633	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	14	reconstruct_from_source	false	https://arxiv.org/pdf/2606.21633v1	cs.LG	2026-06-19T17:36:51	https://arxiv.org/abs/2606.21633v1	HERALD: High-Throughput Block Diffusion LLM Serving via CPU-GPU Cooperative KV Cache Retrieval	t09_efficient_inference_kv_cache	2026-06-19T17:36:51
2606.22082	2606.22082v1	[ "Yifei Wang", "Ruiyin Li", "Peng Liang", "Qiong Feng", "Zengyang Li", "Mojtaba Shahin", "Arif Ali Khan" ]	[ "cs.SE", "cs.AI" ]	arxiv_2606_22082	https://creativecommons.org/licenses/by/4.0/	36	included	true	https://arxiv.org/pdf/2606.22082v1	cs.SE	2026-06-20T14:57:33	https://arxiv.org/abs/2606.22082v1	CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation	t03_software_engineering_code_agents	2026-06-20T14:57:33
2606.22617	2606.22617v1	[ "Hao Vo", "Phu Loc Nguyen", "Khoa Vo", "Sieu Tran", "Duc Minh Nguyen", "Ngo Xuan Cuong", "Nghi D. Q. Bui", "Anh Nguyen", "Duy Minh Ho Nguyen", "Ngan Le" ]	[ "cs.CV" ]	arxiv_2606_22617	https://creativecommons.org/licenses/by/4.0/	14	included	true	https://arxiv.org/pdf/2606.22617v1	cs.CV	2026-06-21T17:47:31	https://arxiv.org/abs/2606.22617v1	OmniSpace: Efficient Geometry Awareness for Autonomous Vehicles MLLMs	t08_3d_spatial_intelligence	2026-06-21T17:47:31
2606.22631	2606.22631v1	[ "Chaoyue Li", "Boxue Yang", "Shengyao Zhou", "Haoyang Wu", "Rui Qian", "Linfeng Zhang" ]	[ "cs.CV" ]	arxiv_2606_22631	https://creativecommons.org/licenses/by/4.0/	28	included	true	https://arxiv.org/pdf/2606.22631v1	cs.CV	2026-06-21T18:33:15	https://arxiv.org/abs/2606.22631v1	4DVLT: Dynamic Scene Understanding with Worldline-Centered Vision-Language Tracking	t08_3d_spatial_intelligence	2026-06-21T18:33:15
2606.22694	2606.22694v1	[ "Danial Kamali", "Tanawan Premsri", "Shreya Rajpal", "Amir Zadeh", "Chuan Li", "Parisa Kordjamshidi" ]	[ "cs.CV", "cs.SC" ]	arxiv_2606_22694	https://creativecommons.org/licenses/by/4.0/	28	included	true	https://arxiv.org/pdf/2606.22694v1	cs.CV	2026-06-21T22:15:48	https://arxiv.org/abs/2606.22694v1	SATURN: Symbolic Spatial Reasoning for Multi-Perspective Grounding	t08_3d_spatial_intelligence	2026-06-21T22:15:48
2606.22813	2606.22813v1	[ "Omar Hashash", "Christo Kurisummoottil Thomas", "Walid Saad", "Merouane Debbah", "Karl Friston", "Adeel Razi" ]	[ "cs.AI" ]	arxiv_2606_22813	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	53	reconstruct_from_source	false	https://arxiv.org/pdf/2606.22813v1	cs.AI	2026-06-22T03:41:31	https://arxiv.org/abs/2606.22813v1	Active Inference as the Test-Time Scaling Law for Physical AI Agents	t06_test_time_scaling_reasoning	2026-06-22T03:41:31
2606.22844	2606.22844v1	[ "Wei Yang", "Bryce Kan", "Shixuan Li", "Li Li", "Yuehan Qin", "Jiate Li", "Paul Bogdan", "Jesse Thomason" ]	[ "cs.AI", "cs.MA" ]	arxiv_2606_22844	https://creativecommons.org/licenses/by/4.0/	36	included	true	https://arxiv.org/pdf/2606.22844v1	cs.AI	2026-06-22T04:41:14	https://arxiv.org/abs/2606.22844v1	RaMem: Contextual Reinstatement for Long-term Agentic Memory	t04_agent_memory_long_horizon	2026-06-22T04:41:14
2606.22906	2606.22906v1	[ "Jiawei He", "Weisong Sun", "Mengyu Shi", "Jie Jia", "Tong Bian", "Xikai Yang", "Dong Sun" ]	[ "cs.SE", "cs.AI" ]	arxiv_2606_22906	https://creativecommons.org/licenses/by/4.0/	12	included	true	https://arxiv.org/pdf/2606.22906v1	cs.SE	2026-06-22T06:44:51	https://arxiv.org/abs/2606.22906v1	From Fragments to Paths: Task-Level Context Recovery for Large Industrial Codebases	t03_software_engineering_code_agents	2026-06-22T06:44:51
2606.22938	2606.22938v1	[ "Stanley Wei", "Juno Kim" ]	[ "cs.LG", "cs.AI" ]	arxiv_2606_22938	https://creativecommons.org/licenses/by/4.0/	38	included	true	https://arxiv.org/pdf/2606.22938v1	cs.LG	2026-06-22T07:16:08	https://arxiv.org/abs/2606.22938v1	Provable Benefits of RLVR over SFT for Reasoning Models: Learning to Backtrack Efficiently	t06_test_time_scaling_reasoning	2026-06-22T07:16:08
2606.23296	2606.23296v1	[ "Chengyu Bai", "Peidong Jia", "Tiecheng Guo", "Yukai Wang", "Rui Ma", "Fangyuan Zhao", "Chunkai Fan", "Xiaobao Wei", "Jintao Chen", "Hao Wang", "Ying Li", "Xiaozhu Ju", "Jian Tang", "Shanghang Zhang" ]	[ "cs.RO" ]	arxiv_2606_23296	https://creativecommons.org/licenses/by/4.0/	16	included	true	https://arxiv.org/pdf/2606.23296v1	cs.RO	2026-06-22T13:09:34	https://arxiv.org/abs/2606.23296v1	IOI: Decoupling Kinematics and Physics for Interactive World Models	t01_embodied_world_models	2026-06-22T13:09:34
2606.23557	2606.23557v1	[ "Jiho Choi", "Seonho Lee", "Seojeong Park", "Hyunjung Shim" ]	[ "cs.CV" ]	arxiv_2606_23557	https://creativecommons.org/licenses/by/4.0/	41	included	true	https://arxiv.org/pdf/2606.23557v1	cs.CV	2026-06-22T16:28:11	https://arxiv.org/abs/2606.23557v1	Dense Reward for Multi-View 3D Reasoning with Global Maps and Local Views	t08_3d_spatial_intelligence	2026-06-22T16:28:11
2606.23610	2606.23610v1	[ "Hongkai Zheng", "Ta-Ying Cheng", "Benjamin Klein", "Yisong Yue", "Zhuoning Yuan" ]	[ "cs.CV" ]	arxiv_2606_23610	https://creativecommons.org/licenses/by-nc-sa/4.0/	24	included	true	https://arxiv.org/pdf/2606.23610v1	cs.CV	2026-06-22T17:11:11	https://arxiv.org/abs/2606.23610v1	Vera: A Layered Diffusion Model for Content-Preserving Video Editing	t07_video_generation_control	2026-06-22T17:11:11
2606.24480	2606.24480v1	[ "Eric Jianfeng Cheng", "Min Hong", "Zhiquan Zeng", "Chuanyu Liu", "Qian Wang", "Melissa Jane Meadowcroft", "Vlad Badilita", "Carlos Miguel Costa", "Senentxu Lanceros-Méndez", "Ying Li", "Chenyao Ma", "Weibo Gong", "Surendra Martha", "Aloysius Soon", "Piao Ma", "Di Zhang", "Teguh Ariy...	[ "cond-mat.mtrl-sci" ]	arxiv_2606_24480	https://creativecommons.org/licenses/by/4.0/	54	included	true	https://arxiv.org/pdf/2606.24480v1	cond-mat.mtrl-sci	2026-06-23T12:16:41	https://arxiv.org/abs/2606.24480v1	Breaking Bottlenecks in Solid Electrolyte Discovery with Large Artificial Intelligence Models	t10_scientific_foundation_models	2026-06-23T12:16:41
2606.24626	2606.24626v1	[ "Chenyang Zhu", "Jiayu Yao", "Kushal Chawla", "Youbing Yin", "Nathan Wolfe", "Pengshan Cai", "Jingyu Wu", "Spencer Hong", "Sangwoo Cho", "Shi-Xiong Zhang", "Daben Liu", "Sambit Sahu", "Erin Babinsky" ]	[ "cs.AI" ]	arxiv_2606_24626	https://creativecommons.org/licenses/by-nc-nd/4.0/	16	reconstruct_from_source	false	https://arxiv.org/pdf/2606.24626v1	cs.AI	2026-06-23T14:23:40	https://arxiv.org/abs/2606.24626v1	SAFARI: Scaling Long Horizon Agentic Fault Attribution via Active Investigation	t04_agent_memory_long_horizon	2026-06-23T14:23:40
2606.26800	2606.26800v1	[ "Kaijun Wang", "Zikai Ouyang", "Xuping Wu", "Jinyi Hong", "Wei Pan", "Haibo Lu", "Jia Pan", "Wei Zhang", "Linfang Zheng" ]	[ "cs.RO" ]	arxiv_2606_26800	https://creativecommons.org/licenses/by-nc-sa/4.0/	8	included	true	https://arxiv.org/pdf/2606.26800v1	cs.RO	2026-06-25T09:38:05	https://arxiv.org/abs/2606.26800v1	SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation	t08_3d_spatial_intelligence	2026-06-25T09:38:05
2606.27876	2606.27876v1	[ "Haoyu Zhang", "Meng Liu", "Qianlong Xiang", "Kun Wang", "Yaowei Wang", "Liqiang Nie" ]	[ "cs.CV", "cs.AI" ]	arxiv_2606_27876	https://arxiv.org/licenses/nonexclusive-distrib/1.0/	10	reconstruct_from_source	false	https://arxiv.org/pdf/2606.27876v1	cs.CV	2026-06-26T09:16:55	https://arxiv.org/abs/2606.27876v1	SpatialUAV: Benchmarking Spatial Intelligence for Low-Altitude UAV Perception, Collaboration, and Motion	t08_3d_spatial_intelligence	2026-06-26T09:16:55

VisDocAgentBench

VisDocAgentBench is a closed-corpus benchmark for visually rich document retrieval. It contains 120 natural-language queries over 2,375 rendered pages from 100 scientific documents. Queries are evenly divided among direct, one-bridge, and two-bridge evidence structures, with one answer page per query.

[Paper] [Code] [Project page]

benchmark/
├── queries.jsonl
├── evaluator_annotations.jsonl
└── topics.json
corpus/
├── documents.jsonl
├── pages.jsonl
└── pages/<document_id>/<page_id>.png
dataset_info.json
LICENSES.md

queries.jsonl contains the query text and topic identifier presented to a retrieval system.
evaluator_annotations.jsonl contains the answer page, evidence level, and ordered latent support pages used by the evaluator and controlled analyses.
documents.jsonl records the exact arXiv version, bibliographic metadata, source URLs, source license, page count, and image-availability status for every document.
pages.jsonl defines all 2,375 page identifiers, document membership, one-based page indices, rendering dimensions, and expected local paths.
topics.json describes the ten corpus topics.

The standard agent harness reads the query text but does not expose answer or support annotations to the planner.

Dataset Statistics

Statistic	Count
Documents	100
Rendered pages	2,375
Queries	120
Direct queries	40
One-bridge queries	40
Two-bridge queries	40
Unique answer pages	120
Directly included page images	1,469
Locally reconstructed page images	906

Loading the Metadata

Each JSONL component has a distinct schema and can be loaded independently:

from datasets import load_dataset

queries = load_dataset(
    "hulx2002/VisDocAgentBench",
    "queries",
    split="test",
)
annotations = load_dataset(
    "hulx2002/VisDocAgentBench",
    "evaluator_annotations",
    split="test",
)
pages = load_dataset(
    "hulx2002/VisDocAgentBench",
    "pages",
    split="corpus",
)

The complete snapshot can be downloaded with the code repository:

python scripts/download_data.py

Reconstructing the Complete Corpus

The repository includes 1,469 rendered page images whose source licenses permit redistribution. The remaining 906 rows remain in corpus/pages.jsonl with image_included=false. The exact source version and PDF URL are recorded in corpus/documents.jsonl.

After cloning the code repository and downloading this dataset into data/, reconstruct the omitted pages with:

python dataset_tools/download_and_render.py
python dataset_tools/validate_dataset.py --require-complete-corpus

The script downloads each specified arXiv version and renders it at 144 DPI. It verifies the expected page count and dimensions before accepting the reconstructed corpus.

Evaluation

Systems return a ranked list of page identifiers or the opaque page handles assigned by the released agent harness. The evaluator reports Recall@1/3/5/10 and MRR@10 overall and by evidence level. Missing or invalid rankings stay in the 120-query denominator and score zero.

Evaluation code, baseline implementations, deterministic preprocessing, and the full agent tool interfaces are available in the code repository. Generated predictions, traces, OCR caches, embeddings, and model weights are not included in this dataset repository.

Licensing

Benchmark-authored queries, evaluator annotations, topics, and corpus metadata are licensed under CC BY 4.0. Source-document page images retain the license of their source document. The license URL and attribution metadata for every source are recorded in corpus/documents.jsonl.

Page images are included only for 55 CC BY 4.0 documents and 4 CC BY-NC-SA 4.0 documents. Sources under the arXiv nonexclusive distribution license or CC BY-NC-ND 4.0 are represented by metadata and local reconstruction instructions, not redistributed images. See LICENSES.md before reusing source pages.

Citation

@article{hu2026visdocagentbench,
  title={VisDocAgentBench: Benchmarking Agents for Visually Rich Document Retrieval},
  author={Hu, Lexiang and Zhang, Yanzhao and Li, Mingxin and Long, Dingkun and Li, Yikang and Zhang, Fuwei and Wang, Yisen and Lin, Zhouchen},
  journal={arXiv preprint arXiv:2608.17889},
  year={2026}
}