Paper Detail

Reliable Knowledge Acquisition for Continual Learning Agents

Yixu Huang, Jiaqi W. Ma

arxiv Score 14.5

Published 2026-10-15 · First seen 2026-10-16

Research Track A · General AI

Abstract

As agents powered by foundation models move into open-ended environments, continual learning from experience becomes essential. For agents built on models with fixed parameters, this learning requires reusing prior task experience in context to improve performance and efficiency on future tasks. Existing memory and experience-learning systems typically focus on either retained facts or procedures, and lack a mechanism for establishing the reliability of claims extracted from the ongoing task stream. Indeed, we find that approximately 12--33\% of entries produced by representative systems harm future task performance when reused. To address these limitations, we propose \textbf{RIKA} (Reliable Knowledge Acquisition), which combines hierarchical extraction of factual, conceptual, procedural, and metacognitive knowledge with Evidence-based Knowledge Admission, which uses source evidence and subsequent experience to govern which claims enter the task agent's context, without additional validation tasks. Across six task streams from Continual Learning Bench and four model backbones, RIKA achieves the highest aggregate score on every backbone.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
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{huang2026reliable,
  title = {Reliable Knowledge Acquisition for Continual Learning Agents},
  author = {Yixu Huang and Jiaqi W. Ma},
  year = {2026},
  abstract = {As agents powered by foundation models move into open-ended environments, continual learning from experience becomes essential. For agents built on models with fixed parameters, this learning requires reusing prior task experience in context to improve performance and efficiency on future tasks. Existing memory and experience-learning systems typically focus on either retained facts or procedures, and lack a mechanism for establishing the reliability of claims extracted from the ongoing task str},
  url = {https://arxiv.org/abs/2610.21739},
  keywords = {cs.CL, cs.AI, cs.LG},
  eprint = {2610.21739},
  archiveprefix = {arXiv},
}

Metadata

{}