WebEye benchmark and Pixel-Searcher agent enable visual perception tasks by using web search to resolve object identities before precise localization or answering.
hub Baseline reference
OneThinker: All-in-one Reasoning Model for Image and Video
Baseline reference. 71% of citing Pith papers use this work as a benchmark or comparison.
20
Pith papers citing it
Baseline
71% of classified citations
abstract
Reinforcement learning (RL) has recently achieved remarkable success in eliciting visual reasoning within Multimodal Large Language Models (MLLMs). However, existing approaches typically train separate models for different tasks and treat image and video reasoning as disjoint domains. This results in limited scalability toward a multimodal reasoning generalist, which restricts practical versatility and hinders potential knowledge sharing across tasks and modalities. To this end, we propose OneThinker, an all-in-one reasoning model that unifies image and video understanding across diverse fundamental visual tasks, including question answering, captioning, spatial and temporal grounding, tracking, and segmentation. To achieve this, we construct the OneThinker-600k training corpus covering all these tasks and employ commercial models for CoT annotation, resulting in OneThinker-SFT-340k for SFT cold start. Furthermore, we propose EMA-GRPO to handle reward heterogeneity in multi-task RL by tracking task-wise moving averages of reward standard deviations for balanced optimization. Extensive experiments on diverse visual benchmarks show that OneThinker delivers strong performance on 31 benchmarks, across 10 fundamental visual understanding tasks. Moreover, it exhibits effective knowledge transfer between certain tasks and preliminary zero-shot generalization ability, marking a step toward a unified multimodal reasoning generalist. All code, model, and data are released.
hub tools
citation-role summary
baseline 3
background 2
dataset 2
citation-polarity summary
representative citing papers
LMMs perceive videos but underexploit visual content for causal reasoning due to textual shortcuts; ProCauEval diagnoses this and ADPO training reduces reliance on priors.
Gen-Searcher is the first trained search-augmented image generation agent using SFT followed by GRPO reinforcement learning with dual text-image rewards, delivering 15-16 point gains on knowledge-intensive benchmarks.
CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
CoLT replaces text-based chain-of-thought in MLLMs with 3-step latent thought chains supervised by a removable external decoder in forward and backward modes, yielding 10.1x faster inference on eight benchmarks.
CRPO applies counterfactual videos and a cross-branch relation reward in RL post-training to reduce shortcut reliance in Video LLMs, with gains shown on the new DyBench paired benchmark.
VersusQ introduces a pairwise margin reasoning framework using large multimodal models to predict signed continuous quality margins between video pairs, claiming improved cross-domain generalization over pointwise scoring methods.
IC-Seg is a multi-turn clarification framework with hierarchical GRPO optimization that resolves ambiguous queries in referring video object segmentation and introduces the Ambi-RVOS benchmark.
Video-Zero is an annotation-free Questioner-Solver co-evolution framework that centers self-evolution on temporally localized evidence to improve video VLMs.
PVM adds a parallel branch to LVLMs that directly supplies visual embeddings to prevent attention decay over long generated sequences, yielding accuracy gains on reasoning tasks with minimal overhead.
CoPD integrates multiple expert capabilities by running parallel RLVR training with bidirectional online policy distillation among experts, outperforming mixed RLVR and sequential OPD while surpassing domain-specific experts on text-image-video reasoning.
Visual replay module and adaptive depth scaling improve multimodal latent reasoning, reaching SOTA benchmarks with faster inference than explicit chain-of-thought methods.
VaLR generates vision-aligned latent tokens before each reasoning step to preserve perceptual cues, improving VSI-Bench accuracy from 33.0% to 52.9%.
Dual Tuning is a data curation method that jointly scores training examples for benefit and for reasoning-gain to choose between reasoning and direct-answer post-training modes for multimodal LLMs.
AdaTooler-V trains MLLMs to adaptively use vision tools via AT-GRPO reinforcement learning and new datasets, reaching 89.8% on V* and outperforming GPT-4o.
Embodied-R1.5 is an 8B EFM achieving SOTA on 16 of 24 embodied VLM benchmarks, fine-tunable to outperform leading VLAs, with claimed zero-shot real-robot generalization.
Chain-of-Glimpse is a reinforcement-learning-based framework that iteratively grounds visual evidence regions to enable multi-step object-aware reasoning in videos.
SimpleSearch-VL improves Qwen3-VL multimodal agent baselines by 15.8-16 points on average using 7K total training examples and reaches parity with Gemini-3-Pro on the 30B variant.
Position paper identifies structural challenges in applying generic agentic AI to Earth Observation and outlines design principles for EO-native agents focused on geospatial state and validity.
EasyVideoR1 delivers an optimized RL pipeline for video understanding in large vision-language models, achieving 1.47x throughput gains and aligned results on 22 benchmarks.
citing papers explorer
-
From Web to Pixels: Bringing Agentic Search into Visual Perception
WebEye benchmark and Pixel-Searcher agent enable visual perception tasks by using web search to resolve object identities before precise localization or answering.
-
Perception Without Engagement: Dissecting the Causal Discovery Deficit in LMMs
LMMs perceive videos but underexploit visual content for causal reasoning due to textual shortcuts; ProCauEval diagnoses this and ADPO training reduces reliance on priors.
-
Gen-Searcher: Reinforcing Agentic Search for Image Generation
Gen-Searcher is the first trained search-augmented image generation agent using SFT followed by GRPO reinforcement learning with dual text-image rewards, delivering 15-16 point gains on knowledge-intensive benchmarks.
-
CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
-
CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts
CoLT replaces text-based chain-of-thought in MLLMs with 3-step latent thought chains supervised by a removable external decoder in forward and backward modes, yielding 10.1x faster inference on eight benchmarks.
-
Learning Spatiotemporal Sensitivity in Video LLMs via Counterfactual Reinforcement Learning
CRPO applies counterfactual videos and a cross-branch relation reward in RL post-training to reduce shortcut reliance in Video LLMs, with gains shown on the new DyBench paired benchmark.
-
VersusQ: Pairwise Margin Reasoning for Generalizable Video Quality Assessment
VersusQ introduces a pairwise margin reasoning framework using large multimodal models to predict signed continuous quality margins between video pairs, claiming improved cross-domain generalization over pointwise scoring methods.
-
Don't Guess, Just Ask: Resolving Ambiguity in Referring Segmentation via Multi-turn Clarification
IC-Seg is a multi-turn clarification framework with hierarchical GRPO optimization that resolves ambiguous queries in referring video object segmentation and introduces the Ambi-RVOS benchmark.
-
Video-Zero: Self-Evolution Video Understanding
Video-Zero is an annotation-free Questioner-Solver co-evolution framework that centers self-evolution on temporally localized evidence to improve video VLMs.
-
Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs
PVM adds a parallel branch to LVLMs that directly supplies visual embeddings to prevent attention decay over long generated sequences, yielding accuracy gains on reasoning tasks with minimal overhead.
-
Co-Evolving Policy Distillation
CoPD integrates multiple expert capabilities by running parallel RLVR training with bidirectional online policy distillation among experts, outperforming mixed RLVR and sequential OPD while surpassing domain-specific experts on text-image-video reasoning.
-
Visual Enhanced Depth Scaling for Multimodal Latent Reasoning
Visual replay module and adaptive depth scaling improve multimodal latent reasoning, reaching SOTA benchmarks with faster inference than explicit chain-of-thought methods.
-
Vision-aligned Latent Reasoning for Multi-modal Large Language Model
VaLR generates vision-aligned latent tokens before each reasoning step to preserve perceptual cues, improving VSI-Bench accuracy from 33.0% to 52.9%.
-
Dual Tuning for Reasoning Efficacy-Driven Data Curation in Multimodal LLM Training
Dual Tuning is a data curation method that jointly scores training examples for benefit and for reasoning-gain to choose between reasoning and direct-answer post-training modes for multimodal LLMs.
-
Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models
Embodied-R1.5 is an 8B EFM achieving SOTA on 16 of 24 embodied VLM benchmarks, fine-tunable to outperform leading VLAs, with claimed zero-shot real-robot generalization.
-
Chain-of-Glimpse: Search-Guided Progressive Object-Grounded Reasoning for Video Understanding
Chain-of-Glimpse is a reinforcement-learning-based framework that iteratively grounds visual evidence regions to enable multi-step object-aware reasoning in videos.
-
SimpleSearch-VL: A Simple Recipe for Multimodal Agentic Deep Search
SimpleSearch-VL improves Qwen3-VL multimodal agent baselines by 15.8-16 points on average using 7K total training examples and reaches parity with Gemini-3-Pro on the 30B variant.
-
Agentic AI for Remote Sensing: Technical Challenges and Research Directions
Position paper identifies structural challenges in applying generic agentic AI to Earth Observation and outlines design principles for EO-native agents focused on geospatial state and validity.
-
EasyVideoR1: Easier RL for Video Understanding
EasyVideoR1 delivers an optimized RL pipeline for video understanding in large vision-language models, achieving 1.47x throughput gains and aligned results on 22 benchmarks.