World models succeed when their latent states are built to meet task-specific sufficiency constraints rather than preserving the maximum amount of information.
Think deep, not just long: Measuring llm reasoning effort via deep-thinking tokens
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
InsightReplay improves LLM accuracy on reasoning benchmarks by extracting and replaying critical insights to maintain their accessibility during extended chain-of-thought generation.
Large reasoning models show measurable hidden-state dynamics that a new statistic can use to distinguish correct reasoning trajectories without labels.
citing papers explorer
-
Latent State Design for World Models under Sufficiency Constraints
World models succeed when their latent states are built to meet task-specific sufficiency constraints rather than preserving the maximum amount of information.
-
Stateful Reasoning via Insight Replay
InsightReplay improves LLM accuracy on reasoning benchmarks by extracting and replaying critical insights to maintain their accessibility during extended chain-of-thought generation.
-
Spatiotemporal Hidden-State Dynamics as a Signature of Internal Reasoning in Large Language Models
Large reasoning models show measurable hidden-state dynamics that a new statistic can use to distinguish correct reasoning trajectories without labels.