A sleep mechanism with N offline recurrent passes consolidates context into fast weights, improving performance on reasoning tasks where standard transformers fail.
hub
Learning by distilling context
23 Pith papers cite this work. Polarity classification is still indexing.
23
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
Self-distillation token rewards measure input-response-feedback pointwise mutual information, and CREDIT extracts the input-specific component with contrastive baselines to improve LLM reasoning performance.
Distillation signals align better with ideal updates on incorrect student rollouts than correct ones, with optimal teacher context depending on student capacity and task.
CoDistill-GRPO lets small and large models mutually improve via co-distillation in GRPO, raising small-model math accuracy by over 11 points while cutting large-model training time by about 18%.
NPO uses a policy's own near-future checkpoint as auxiliary trajectories to maximize effective learning signal S = Q/V, improving performance from 57.88 to 63.15 on Qwen3-VL-8B-Instruct with GRPO while accelerating convergence.
LLMs trained on simple specification gaming generalize to zero-shot reward tampering including rewriting their own reward function.
D2D distills distributional shifts between a suspected model and its base into a cartridge adapter to amplify and detect stealth biases in LLMs across multiple types.
HMARS introduces a hierarchical multi-agent memory system that outperforms standard retrieval and other baselines on long-document and multi-turn reasoning tasks through improved evidence coverage.
ThinkSwitch uses iterative self-distillation with QLoRA and spherical weight interpolation to raise both instruct and thinking checkpoint accuracy on small AIME and PubMedQA sets using only 15 human prompts per domain.
DASD improves math reasoning in LLMs by adaptively directing self-distillation based on per-token entropy to balance exploration and step accuracy, outperforming prior self-distillation and RLVR baselines on six benchmarks.
Attention-state memory externalizes long prefixes into a lightweight lookup table of precomputed attention states, yielding higher accuracy than standard in-context learning at fixed memory budgets and lower latency than full attention.
SSOPD converts intra-group correct-wrong contrast into process supervision by distilling a teacher distribution from the shortest correct completion into prefixes of the longest wrong completion, improving GRPO on AIME and HMMT benchmarks.
VSPO samples rollouts at varying steering intensities to improve behavioral control in LLMs while preserving task accuracy.
Reasoning language models extract answers from sparse, order-shuffled chain-of-thought traces with little accuracy loss.
TSUBASA improves long-horizon personalization in LLMs via dynamic memory evolution for writing and context-distillation self-learning for reading, outperforming Mem0 and Memory-R1 on Qwen-3 benchmarks while reducing token use.
Training LLMs on data that enforces priority levels for instructions makes models robust to prompt injection attacks, including unseen ones, with little loss on standard tasks.
A 540B-parameter LLM improves reasoning performance on GSM8K, DROP, OpenBookQA, and ANLI-A3 by fine-tuning on self-generated high-confidence CoT solutions from unlabeled data.
A single LLM rewrite of skill descriptions using false positive and negative cases matches manual optimization performance in production, with most other pipeline components adding little value.
HyperLoRA amortizes federated LoRA adaptation via hypernetwork-generated initializations and product-space aggregation to fix structural bias and initialization lag.
Existing methods for turning LLM interaction experience into parametric skills collapse over multiple iterations; principle-level experience, step-wise injection, and off-policy teacher distillation yield more stable continual learning.
LLM post-training is unified as off-policy or on-policy interventions that expand support for useful behaviors, reshape policies within reachable states, or consolidate behavior across training stages.
Two-stage fine-tuning of Qwen2.5-VL-32B improves success rates on single-click web tasks from 86% to 94%.
SELFCI uses complementary self-distillation with two reverse KL divergences to align LLMs to contextual integrity while preserving utility, outperforming RL baselines like GRPO in agentic settings.
citing papers explorer
-
From Generic Correlation to Input-Specific Credit in On-Policy Self Distillation
Self-distillation token rewards measure input-response-feedback pointwise mutual information, and CREDIT extracts the input-specific component with contrastive baselines to improve LLM reasoning performance.
-
Near-Future Policy Optimization
NPO uses a policy's own near-future checkpoint as auxiliary trajectories to maximize effective learning signal S = Q/V, improving performance from 57.88 to 63.15 on Qwen3-VL-8B-Instruct with GRPO while accelerating convergence.
-
Large Language Model Post-Training: A Unified View of Off-Policy and On-Policy Learning
LLM post-training is unified as off-policy or on-policy interventions that expand support for useful behaviors, reshape policies within reachable states, or consolidate behavior across training stages.