Data2Story is a multi-agent framework that generates evidence-grounded multimodal articles from data, evaluated on 18 articles against human pieces for verifiability, angle coverage, and quality across human, rubric, and automated judges.
Mixed citations
arXiv preprint arXiv:2409.12959 , year=
Mixed citation behavior. Most common role is background (50%).
23
Pith papers citing it
Background
50% of classified citations
citation-role summary
background 4
dataset 2
citation-polarity summary
verdicts
UNVERDICTED 23representative citing papers
PixelRAG shows that operating RAG entirely over web screenshots outperforms text-based retrieval on NQ, SimpleQA, MMSearch, LiveVQA, and MoNaCo, with up to 18.1% accuracy gains and 3x token savings via image compression.
SVFSearch is the first open benchmark for short-video frame search in the Chinese gaming domain, providing a frozen retrieval environment and showing performance gaps of 13-29 points between direct QA models, practical agents, and oracle knowledge.
FIKA-Bench is a leakage-aware benchmark of 311 instances showing that even the best large multimodal models and tool-equipped agents reach only 25.1% accuracy on fine-grained recognition questions that require external evidence search and verification.
WebEye benchmark and Pixel-Searcher agent enable visual perception tasks by using web search to resolve object identities before precise localization or answering.
ZipRerank delivers state-of-the-art multimodal listwise reranking accuracy for long documents at up to 10x lower latency via early interaction and single-pass scoring.
VISOR is a unified agentic VRAG framework with Evidence Space structuring, visual action evaluation/correction, and dynamic sliding-window trajectories trained via GRPO-based RL that achieves SOTA performance on long-horizon visual reasoning benchmarks.
VideoDR is a new benchmark for open-web video deep research that tests multimodal models on cross-frame visual anchor extraction, interactive retrieval, and multi-hop reasoning over joint video-web evidence.
IMAgent trains a multi-image vision agent via pure end-to-end RL with visual reflection tools and a two-layer motion trajectory masking strategy, reaching SOTA on single- and multi-image benchmarks while revealing tool-use effects on attention.
MMSearch-R1 uses reinforcement learning to train multimodal models for on-demand multi-turn internet search with image and text tools, outperforming same-size RAG baselines and matching larger ones while cutting search calls by over 30%.
A decoupled training-free IBA framework for KB-VQA selects entities via MLLM candidate choice then ranks evidence with off-the-shelf re-rankers, outperforming coupled fine-tuned baselines on Encyclopedic-VQA and InfoSeek.
TAPO corrects credit misassignment in RL for multimodal search agents by using tool parameter similarity to share advantages across equivalent actions.
AXPO addresses the Thinking-Acting Gap in agentic RL training by targeted resampling of tool calls in all-wrong subgroups, delivering +1.8pp gains over GRPO on nine multimodal benchmarks with an 8B model beating a 32B baseline on Pass@4.
HyperEyes presents a parallel multimodal search agent using dual-grained efficiency-aware RL with a new TRACE reward and IMEB benchmark, claiming 9.9% higher accuracy and 5.3x fewer tool calls than prior open-source agents.
DR-MMSearchAgent derives batch-wide trajectory advantages and uses differentiated Gaussian rewards to prevent premature collapse in multimodal agents, outperforming MMSearch-R1 by 8.4% on FVQA-test.
LMM-Searcher uses file-based visual UIDs and a fetch tool plus 12K synthesized trajectories to fine-tune a multimodal agent that scales to 100-turn horizons and reaches SOTA among open-source models on MM-BrowseComp and MMSearch-Plus.
DeepEyesV2 uses a two-stage cold-start plus reinforcement learning pipeline to produce an agentic multimodal model that adaptively invokes tools and outperforms direct RL on real-world reasoning benchmarks.
WebWatcher introduces a vision-language deep research agent trained on synthetic multimodal trajectories and RL that outperforms baselines on VQA benchmarks, along with a new BrowseComp-VL evaluation.
MathVis-Fine proposes a dataset with fine-grained visual annotations and dependency ratings plus a progressive two-stage training paradigm to align visual supervision with sample-specific necessity in multimodal mathematical reasoning.
Proposes image-bank harness and ODE closed-loop data generation to boost multimodal deep search agents, reporting average score gains from 24.9% to 39.0% on 8 benchmarks for 8B model and 30.6% to 41.5% for 30B.
GLM-5V-Turbo integrates multimodal perception as a core part of reasoning and execution for agentic tasks, reporting strong results in visual tool use and multimodal coding while keeping text-only performance competitive.
A sandbox-trained multimodal search agent with process-oriented rewards transfers zero-shot to real Google Search and outperforms prior methods on FVQA, InfoSeek, and MMSearch.
Valley3 is an omni MLLM for e-commerce that uses a four-stage pre-training pipeline plus post-training for controllable reasoning and agentic search, outperforming baselines on e-commerce benchmarks while staying competitive on general ones.
citing papers explorer
-
Ground Then Rank: Revisiting Knowledge-Based VQA with Training-Free Entity Identification
A decoupled training-free IBA framework for KB-VQA selects entities via MLLM candidate choice then ranks evidence with off-the-shelf re-rankers, outperforming coupled fine-tuned baselines on Encyclopedic-VQA and InfoSeek.
-
Agent Explorative Policy Optimization for Multimodal Agentic Reasoning
AXPO addresses the Thinking-Acting Gap in agentic RL training by targeted resampling of tool calls in all-wrong subgroups, delivering +1.8pp gains over GRPO on nine multimodal benchmarks with an 8B model beating a 32B baseline on Pass@4.
-
Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents
Proposes image-bank harness and ODE closed-loop data generation to boost multimodal deep search agents, reporting average score gains from 24.9% to 39.0% on 8 benchmarks for 8B model and 30.6% to 41.5% for 30B.