Showing cs.CLShow all
3 papers · 1 filter
cs.CL2026
LLMs Encode Their Failures: Predicting Success from Pre-Generation Activations
William Lugoloobi, Thomas Foster, William Bankes +1
Running LLMs with extended reasoning on every problem is expensive, but determining which inputs actually require additional compute remains challenging. We investigate whether the…
cs.CL2026
Task-Specific Knowledge Distillation via Intermediate Probes
Ryan Brown, Chris Russell
Knowledge distillation from large language models (LLMs) assumes that the teacher's output distribution is a high-quality training signal. On reasoning tasks, this assumption is fr…
cs.CL2025
LLMs Encode How Difficult Problems Are
William Lugoloobi, Chris Russell
Large language models exhibit a puzzling inconsistency: they solve complex problems yet frequently fail on seemingly simpler ones. We investigate whether LLMs internally encode pro…