LLM Digest

Story

arxiv_llm_reliability · Sep 27, 2026 · paper

Source brief

Trajectory Unlearning on LLM-based Agents

arxiv.orgSep 27, 2026
original source linked

In brief

Existing large language model (LLM) unlearning has focused primarily on removing specific knowledge, such as harmful facts, private data, or copyrighted content. However, as LLMs are increasingly deployed as autonomou...

Feed lens

agenticeval

Read the original at arxiv.org →Open in live feed Read that day’s brief

Trajectory Unlearning on LLM-based Agents

Earlier in this thread 4 items

SpaCellAgent: A Self-Evolving LLM-Based Multi-Agent Framework for Trajectory Analysis

Behavior2Trip: Towards Personalized Travel Planning via User Behavior Trajectory

LSEAD: A Privacy-Preserving LLM-Based Speech Analysis Framework for Early Alzheimer's Disease Screening

Dependency-Aware Trajectory Refinement for Efficient Multi-Turn Agent Fine-Tuning