DASH assigns segment-level credit in reasoning traces using drift toward ground-truth answers, yielding 50.8% accuracy on AIME25 versus 45.4% for GRPO while reducing overthinking behaviors.
Emily Pronin, Daniel Y Lin, and Lee Ross
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
years
2026 7representative citing papers
DynaSteer is a dynamic representation editing framework that uses pattern clustering, Fisher-LDA, and lookahead entropy monitoring to steer LLM reasoning trajectories toward truth on MATH and coding tasks.
Post-training quantization increases overthinking errors in reasoning models; a logit penalty on curated overthinking markers reduces CoT length 12-23% without accuracy loss.
GitHub Actions workflows achieve only 28% overall compliance with best practices, with LLMs enabling an 81% reduction in verification effort via hybrid adjudication but still requiring expert oversight for security judgments.
RMR damps the most persistent directions in the Transformer value cache, reducing LLM mode collapse as measured by correlation dimension.
Reasoning SFT generalizes cross-domain conditionally on sufficient optimization, high-quality long-CoT data, and strong base models, while degrading safety.
An exploratory red-teaming study documents eleven cases of security, privacy, and governance failures in autonomous language-model agents with tool access and persistent memory.
citing papers explorer
-
Know When to Stop: Segment-Level Credit Assignment for Reducing Overthinking
DASH assigns segment-level credit in reasoning traces using drift toward ground-truth answers, yielding 50.8% accuracy on AIME25 versus 45.4% for GRPO while reducing overthinking behaviors.
-
Search for Truth from Reasoning: A Dynamic Representation Editing Framework for Steering LLM Trajectories
DynaSteer is a dynamic representation editing framework that uses pattern clustering, Fisher-LDA, and lookahead entropy monitoring to steer LLM reasoning trajectories toward truth on MATH and coding tasks.
-
Quantized Reasoning Models Think They Need to Think Longer, but They Do Not
Post-training quantization increases overthinking errors in reasoning models; a logit penalty on curated overthinking markers reduces CoT length 12-23% without accuracy loss.
-
How Compliant Are GitHub Actions Workflows? A Checklist-Based Study with LLM-Assisted Auditing
GitHub Actions workflows achieve only 28% overall compliance with best practices, with LLMs enabling an 81% reduction in verification effort via hybrid adjudication but still requiring expert oversight for security judgments.
-
Escaping Mode Collapse in LLM Generation via Geometric Regulation
RMR damps the most persistent directions in the Transformer value cache, reducing LLM mode collapse as measured by correlation dimension.
-
Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability
Reasoning SFT generalizes cross-domain conditionally on sufficient optimization, high-quality long-CoT data, and strong base models, while degrading safety.
-
Agents of Chaos
An exploratory red-teaming study documents eleven cases of security, privacy, and governance failures in autonomous language-model agents with tool access and persistent memory.