Toward Counterfactual Learning for Selective Re-Examination in Agentic Visual Search
cs.CV
Submitted: 2026-10-31
Updated: 2026-10-31
Terminology
Sources
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- VisualNeedle: Benchmarking Active Visual Search in Information-Dense Scenes
- Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
- TACO: Tool-Augmented Credit Optimization for Agentic Tool Use
- Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs
- VistaHop: Benchmarking Long-Horizon Visual DeepSearch
- Entropy Is Not Enough: Unlocking Effective Reinforcement Learning for Visual Reasoning via Vision-Anchored Token Selection
- VinePPO: Refining Credit Assignment in RL Training of LLMs
- OtterHD: A High-Resolution Multi-modality Model
- Hierarchical Budget Policy Optimization for Adaptive Reasoning
- What Does Vision Tool-Use Reinforcement Learning Really Learn? Disentangling Tool-Induced and Intrinsic Effects for Crop-and-Zoom
- Learning Which Evidence-Seeking Decisions Matter: Calibrated Counterfactual Replay for Language Agents
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Reinforcing Multi-Turn Reasoning in LLM Agents via Turn-Level Reward Design
- Nice Fold or Hero Call: Learning Budget-Efficient Thinking under Policy-Dependent Solvability
- InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs