Reasoning Shift: How Context Silently Shortens LLM Reasoning
Paper • 2604.01161 • Published • 31
Research to build upon
-
-
Steerable Visual Representations
Paper • 2604.02327 • Published • 53 -
The Depth Ceiling: On the Limits of Large Language Models in Discovering Latent Planning
Paper • 2604.06427 • Published • 11 -
BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation
Paper • 2604.09497 • Published • 28 -
Masked by Consensus: Disentangling Privileged Knowledge in LLM Correctness
Paper • 2604.12373 • Published • 8 -
GlotOCR Bench: OCR Models Still Struggle Beyond a Handful of Unicode Scripts
Paper • 2604.12978 • Published • 5 -
Target Policy Optimization
Paper • 2604.06159 • Published • 22 -
IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures
Paper • 2604.07709 • Published • 1