TACO soft-suppresses positive GRPO credit on high tail-risk tokens (surprisal above local entropy) and consistently beats GRPO-style baselines on three LLMs and eight reasoning benchmarks while stabilizing long training.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
years
2026 3roles
dataset 1polarities
use dataset 1representative citing papers
SORT turns all-wrong prompts into selective learning signals by weighting tokens more predictable under plan guidance from reference solutions, improving over GRPO on reasoning benchmarks especially for weaker models.
ExpThink reduces average CoT response length by up to 77% while improving accuracy on math benchmarks via experience-guided reward shaping and difficulty-adaptive advantage in RL.
citing papers explorer
-
When Implausible Tokens Get Reinforced: Tail-Aware Credit Calibration for LLM Reinforcement Learning
TACO soft-suppresses positive GRPO credit on high tail-risk tokens (surprisal above local entropy) and consistently beats GRPO-style baselines on three LLMs and eight reasoning benchmarks while stabilizing long training.
-
Selective Off-Policy Reference Tuning with Plan Guidance
SORT turns all-wrong prompts into selective learning signals by weighting tokens more predictable under plan guidance from reference solutions, improving over GRPO on reasoning benchmarks especially for weaker models.
-
ExpThink: Experience-Guided Reinforcement Learning for Adaptive Chain-of-Thought Compression
ExpThink reduces average CoT response length by up to 77% while improving accuracy on math benchmarks via experience-guided reward shaping and difficulty-adaptive advantage in RL.