POW3R adapts rubric criterion weights via rollout contrast in RLVR to improve mean reward, strict completion rates, and training speed over static rubric aggregation on multimodal and text tasks.
Perceptual-evidence anchored reinforced learning for multimodal reasoning
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
baseline 1
citation-polarity summary
years
2026 4verdicts
UNVERDICTED 4representative citing papers
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.
Echo-α integrates organ-specific detectors with global visual context via an invoke-and-reason agentic loop, trained on a nine-task curriculum plus sequential RL, to achieve superior grounding (56.73%/43.78% F1@0.5) and diagnosis (74.90%/49.20% accuracy) on cross-center renal and breast ultrasound.
The paper introduces the Proxy Compression Hypothesis as a unifying framework explaining reward hacking in RLHF as an emergent result of compressing high-dimensional human objectives into proxy reward signals under optimization pressure.
citing papers explorer
-
Not Every Rubric Teaches Equally: Policy-Aware Rubric Rewards for RLVR
POW3R adapts rubric criterion weights via rollout contrast in RLVR to improve mean reward, strict completion rates, and training speed over static rubric aggregation on multimodal and text tasks.
-
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.
-
Echo-{\alpha}: Large Agentic Multimodal Reasoning Model for Ultrasound Interpretation
Echo-α integrates organ-specific detectors with global visual context via an invoke-and-reason agentic loop, trained on a nine-task curriculum plus sequential RL, to achieve superior grounding (56.73%/43.78% F1@0.5) and diagnosis (74.90%/49.20% accuracy) on cross-center renal and breast ultrasound.
-
Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
The paper introduces the Proxy Compression Hypothesis as a unifying framework explaining reward hacking in RLHF as an emergent result of compressing high-dimensional human objectives into proxy reward signals under optimization pressure.