arXiv AI

Making Every Tool Call Count: Necessary Tool-Evidence Path Rewards for Agentic Vision-Language Models

The paper introduces the Necessary Tool‑Evidence Path (NTEP) annotation scheme and its associated reward mechanism (NTEP‑R) to better supervise vision‑language models that use external tools. By explicitly specifying which evidence is needed and penalizing redundant tool calls, the authors train an 8B‑parameter model that shows improved accuracy and tool‑use efficiency across seven image‑grounded benchmarks. The approach demonstrates that fine‑grained supervision of tool‑evidence paths is essential for robust agentic VLM performance.

Hugging Face Trending Papers
Jul 23

CRAG-MM-Diagnostics: Enabling Stage-Wise Analysis of Knowledge-Intensive VQA

Knowledge-Intensive Visual Question Answering (KI-VQA) benchmarks evaluate Vision-Language Models (VLMs) as multimodal knowledge assistants by requiring external information beyond a provided image to answer questions. KI-VQA involves multiple sub-problems -referring expression understanding, visual grounding, object recognition, knowledge retrieval, and reasoning-yet existing benchmarks typically report only end-task accuracy, obscuring where failures arise.

arXiv AI
Aug 25

ATP-Bench: Towards Agentic Tool Planning for MLLM Interleaved Generation

ATP‑Bench proposes a new benchmark for evaluating agentic tool planning in multimodal large language models (MLLMs) that generate interleaved text-and-image responses. The benchmark contains 7,702 QA pairs, including 1,592 visual‑question‑answer pairs, across eight categories and 25 visual‑critical intents, all verified by humans. A Multi‑Agent MLLM‑as‑a‑Judge (MAM) system is introduced to assess tool‑call precision, missed opportunities, and overall response quality without relying on ground‑truth references.

By Yinuo Liu, Zi Qian, Heng Zhou, Jiahao Zhang, Yajie Zhang, Zhihang Li, Mengyu Zhou, Erchao Zhao, Xiaoxi Jiang, Guanjun Jiang
arXiv AI
Jul 24

CRAG-MM-Diagnostics: Enabling Stage-Wise Analysis of Knowledge-Intensive VQA

arXiv:2607. 21155v1 Announce Type: cross Abstract: Knowledge-Intensive Visual Question Answering (KI-VQA) benchmarks evaluate Vision-Language Models (VLMs) as multimodal knowledge assistants by requiring external information beyond a provided image to answer questions.

By Hanseok Oh, Parishad BehnamGhader, Benno Krojer, Hyunji Lee, Paul Liang, Siva Reddy, Verna Dankers
arXiv AI
Jul 21

Seeing What Is Actually There: PriVE-Bench and PriVE-Tools for Counterfactual Evaluation of Agentic Visual Evidence in VLMs

arXiv:2607. 16311v1 Announce Type: cross Abstract: Vision-language models (VLMs) often answer visual questions using learned language and category priors rather than grounding their predictions in the image itself.

By Jingyu Sun, Jiachen Tu, Yuyang Xue, Yaoxin Jiang, Guoyi Xu, Zhengtao Yao, Rui Qian, Yizheng Sun, Hongpeng Zhou, Jingyuan Sun, Yan Lin
arXiv Computer Vision
Sep 1

Learning to Search: A Decision-Based Agent for Knowledge-Based Visual Question Answering

arXiv:2604.07146v3 Announce Type: replace Abstract: Knowledge-based visual question answering (KB-VQA) requires vision-language models to understand images and use external knowledge, especially for...

By Zhuohong Chen, Zhenxian Wu, Yunyao Yu, Hangrui Xu, Zirui Liao, Zhifang Liu, Xiangwen Deng, Pen Jiao, Haoqian Wang
arXiv AI
Sep 1

WeAgent-MMSearch: Native Text-Vision Interaction for Multimodal Search Agents

WeAgent-MMSearch introduces a multimodal search agent that preserves retrieved images as persistent references, enabling the model to inspect, process, and cite them throughout a search trajectory. The system includes a harness (WeAgent-Harness), a post‑training method (FA‑GSPO) that recovers salvageable rollouts, and a new benchmark (VisTarget‑Bench) to evaluate image‑retrieval versus visual‑perception failures. Evaluation shows that agentic post‑training boosts performance by 19.22 points, allowing the model to outperform similarly sized open‑source models and compete with much larger ones.

By Zongkai Liu, Hui Zhang, Liqiang Niu, Zhen Cao, Han Li, Juntao Liu, Wenchao Chen, Chengduo Zhao, Chao Yu, Fandong Meng
arXiv Computer Vision
Aug 27

AdaVDR: Adaptive Tool Use and Reflection for Video Deep Research

AdaVDR is an adaptive video deep research agent that selects and reflects on tool usage based on the task and the model’s capabilities. It constructs a specialized data pipeline to generate high‑quality QA pairs and uses model‑conditioned filtering to remove unnecessary tool calls. The agent is trained with supervised fine‑tuning and reinforcement learning, achieving top performance on the VDR‑EE benchmark and significant gains on VideoDR.

By Xintong Zhang, Xiaomeng Fan, Shilin Yan, Ekko He, Zicheng Liu, Zijian Zou, Guannan Zhang, Yuwei Wu, Zhi Gao, Hongwei Xue
arXiv AI
Jul 31

VistaHop: Benchmarking Long-Horizon Visual DeepSearch

arXiv:2606. 03273v2 Announce Type: replace-cross Abstract: Visual DeepSearch tasks require multimodal large language models (MLLMs) to resolve complex visual queries by repeatedly inspecting image regions, grounding reasoning in visual evidence, and connecting fine-grained clues across multiple steps.

By Hang He, Chuhuai Yue, Chengqi Dong, Chengcheng Wan, Ting Su, Haiying Sun, Jiajun Chai, Xiaohan Wang, Guojun Yin
arXiv AI
Aug 26

VideoHarness-RSI: Recursive Harness Self-Improvement for Long-Video Understanding with Frozen Vision-Language Models

VideoHarness‑RSI explores how improving the executable context‑construction program alone can enhance long‑video understanding with frozen vision‑language models. By recursively searching for better harnesses—programs that select and structure video segments—using an outer‑loop proposer that learns from prior programs and execution traces, the method consistently outperforms weaker hand‑crafted baselines and further improves upon stronger ones. The resulting harnesses transfer to other long‑video benchmarks without additional search, demonstrating that executable context construction is a distinct, reusable optimization layer.

By Guoyang Xu, Hao Chen