PrisMem：智能體記憶的能力驅動式自我演化
Capability-Driven Self-Evolution of Agent Memory
研究團隊提出 PrisMem，將智能體記憶程式的演化搜尋由整體表現擴展至個別能力維度，並以依賴感知的能力選擇、歷史引導診斷及軌跡引導整合改進記憶程式。在 BEAM-1M 和 LongMemEval-M 上，PrisMem 分別較最強基線高 10.54 和 7.83 個百分點，並在 million-token 歷史記錄上展現成效。
Published on Oct 5
Authors:
,
,
,
,
,
,
,
,
,
Abstract
Memory self-evolution uses task feedback to iteratively improve executable memory programs that store and retrieve information from past interactions. Existing approaches typically adopt holistic evolution, deriving revision directions from mixed feedback and judging progress by overall performance. This can obscure optimization directions and hide capability-specific gains offset by regressions elsewhere, leaving promising directions underexplored. We introduce capability-driven evolution, which extends search guidance from overall performance to individual capability dimensions, preserving promising revisions and expanding exploration beyond the boundaries of holistic evolution. We propose PrisMem, which uses dependency-aware capability selection to prioritize targets with potential cross-capability benefits and history-guided diagnosis to refine capability specialists. Trace-guided integration compares evaluated programs on paired differential cases, using their behavioral differences to consolidate complementary gains into a unified memory program. Experiments show that PrisMem outperforms the strongest baselines by 10.54 and 7.83 percentage points on BEAM-1M and LongMemEval-M, respectively, demonstrating its effectiveness on million-token histories.
View arXiv page View PDF Project page Add to collection
Get this paper in your agent:
hf papers read 2610.06361
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.06361 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.06361 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.06361 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源：HuggingFace Daily Papers（社區熱門論文） · huggingface.co