Watch Before You Answer: Learning from Visually Grounded Post-Training
Paper • 2604.05117 • Published • 26
Yury Panikov
panikov
AI & ML interests
None yet
Recent Activity
upvoted a paper about 6 hours ago
Watch Before You Answer: Learning from Visually Grounded Post-Training upvoted a paper about 6 hours ago
Vanast: Virtual Try-On with Human Image Animation via Synthetic Triplet Supervision upvoted a paper about 6 hours ago
Beyond Accuracy: Unveiling Inefficiency Patterns in Tool-Integrated ReasoningOrganizations
None yet
upvoted 15
papers about 6 hours ago
Vanast: Virtual Try-On with Human Image Animation via Synthetic Triplet Supervision
Paper • 2604.04934 • Published • 32
Beyond Accuracy: Unveiling Inefficiency Patterns in Tool-Integrated Reasoning
Paper • 2604.05404 • Published • 33
ThinkTwice: Jointly Optimizing Large Language Models for Reasoning and Self-Refinement
Paper • 2604.01591 • Published • 32
GBQA: A Game Benchmark for Evaluating LLMs as Quality Assurance Engineers
Paper • 2604.02648 • Published • 38
ACES: Who Tests the Tests? Leave-One-Out AUC Consistency for Code Generation
Paper • 2604.03922 • Published • 48
Learning to Retrieve from Agent Trajectories
Paper • 2604.04949 • Published • 57
Claw-Eval: Toward Trustworthy Evaluation of Autonomous Agents
Paper • 2604.06132 • Published • 99
Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding
Paper • 2604.05015 • Published • 205
BidirLM: From Text to Omnimodal Bidirectional Encoders by Adapting and Composing Causal LLMs
Paper • 2604.02045 • Published • 19
Mimic Intent, Not Just Trajectories
Paper • 2602.08602 • Published • 9
Synthetic Sandbox for Training Machine Learning Engineering Agents
Paper • 2604.04872 • Published • 8
Scaling Teams or Scaling Time? Memory Enabled Lifelong Learning in LLM Multi-Agent Systems
Paper • 2604.03295 • Published • 7
Type-Checked Compliance: Deterministic Guardrails for Agentic Financial Systems Using Lean 4 Theorem Proving
Paper • 2604.01483 • Published • 4
Locally Confident, Globally Stuck: The Quality-Exploration Dilemma in Diffusion Language Models
Paper • 2604.00375 • Published • 3
commented a
paper 1 day ago
upvoted 4
papers 1 day ago
LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models
Paper • 2603.28301 • Published • 75
The Geometric Alignment Tax: Tokenization vs. Continuous Geometry in Scientific Foundation Models
Paper • 2604.04155 • Published • 6
Cog-DRIFT: Exploration on Adaptively Reformulated Instances Enables Learning from Hard Reasoning Problems
Paper • 2604.04767 • Published • 3
Emergent Compositional Communication for Latent World Properties
Paper • 2604.03266 • Published • 3