Paper Detail

Different Feedback, Different Updates: Selective Self-Learning from User Interactions for Large Language Models

Xuanchen Li, Haitao Li, Yujia Zhou, Qingyi Pan, Heng Wang, Yiqun Liu, Min Zhang, Qingyao Ai

arxiv Score 9.2

Published 2026-08-10 · First seen 2026-08-11

General AI

Abstract

User feedback offers natural supervision for persistent LLM improvement, but a single message may support multiple behavioral changes with different scopes of generalization. We introduce SLIFT, a selective self-learning framework built on a task-relative view of user feedback. SLIFT decomposes each feedback message into atomic components and interprets each component relative to the original task as Fix, Spec, or Null: requirements for task validity, compatible condition-specific refinements, or content with no reliable positive update direction. To incorporate each change at the appropriate scope, SLIFT trains two complementary LoRA adapters on a shared frozen backbone: a Generalist that consolidates Fix requirements into default behavior through feedback-conditioned self-distillation, and a Specialist that observes only the task and Generalist response to supply residual guidance for applicable, unmet Spec refinements. Null components induce no positive update. Across backbones, SLIFT achieves strong performance on both MemoryBench and WildFB, with targeted analyses further examining its underlying mechanisms. We release our code at https://anonymous.4open.science/r/SLIFT.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{li2026different,
  title = {Different Feedback, Different Updates: Selective Self-Learning from User Interactions for Large Language Models},
  author = {Xuanchen Li and Haitao Li and Yujia Zhou and Qingyi Pan and Heng Wang and Yiqun Liu and Min Zhang and Qingyao Ai},
  year = {2026},
  abstract = {User feedback offers natural supervision for persistent LLM improvement, but a single message may support multiple behavioral changes with different scopes of generalization. We introduce SLIFT, a selective self-learning framework built on a task-relative view of user feedback. SLIFT decomposes each feedback message into atomic components and interprets each component relative to the original task as Fix, Spec, or Null: requirements for task validity, compatible condition-specific refinements, o},
  url = {https://arxiv.org/abs/2608.09109},
  keywords = {cs.AI},
  eprint = {2608.09109},
  archiveprefix = {arXiv},
}

Metadata

{}