Entity · paper

Notes to Self: Can LLMs Benefit from Experiential Abstractions?

paperactiveprovisionalnotes-to-self-can-llms-benefit-from-experiential-abstractions--efb76c37·1 events·first seen 4d ago

Aliases: Notes to Self: Can LLMs Benefit from Experiential Abstractions?

Co-occurring entities

MATH

More like this (12)

LLM-as-a-Coach: Experiential Learning for Non-Verifiable Tasks Human Adults and LLMs as Scientists: Who Benefits from Active Exploration?Metacognition in LLMs: Foundations, Progress, and Opportunities Metacognition in LLMs: Foundations, Progress, and Opportunities From Plausible to Actionable: A Position on LLM Self-Explanations Can LLMs Judge Better Than They Generate? Evaluating Task Asymmetry, Mechanistic Interpretability and Transferability for In-Context QA Teaching LLMs to Self-Evolve: Cultivating Core Meta-Skills with Reinforcement Learning The Masked Advantage: Uncovering Local-Language Access to Cultural Knowledge in LLMs Do LLM Embedding Spaces Recover Expert Structure?LLM inference Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?

Recent events (1)

6arXiv · cs.CL·4d ago·source ↗

LLMs can extract and reuse experiential abstractions to improve reasoning, study finds

Researchers investigate whether LLMs can distill solution traces into reusable natural-language abstractions (strategies, reminders) and apply them to future problems, analogous to human experiential learning. Using the MATH training set, abstractions are extracted by either a stronger teacher model or the LLMs themselves into a retrievable library, then applied via inference-time retrieval or RL with abstraction-augmented prompts. Both modes improve performance on mathematical and logical reasoning benchmarks, with self-extracted abstractions matching teacher-extracted ones. The framework generalizes across datasets and models, suggesting a promising direction for continual self-improvement in LLMs.

Evaluation and Benchmarking Agent and Tool Ecosystem MATH Notes to Self: Can LLMs Benefit from Experiential Abstractions?+1 more