arXiv AI
Aug 12
Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding
arXiv:2608. 11095v1 Announce Type: new Abstract: Agentic coding READMEs like CLAUDE.
Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding
Agentic coding READMEs like CLAUDE. md grow without bound in real repositories, stopping only when the repository retires or someone rewrites the file wholesale.
Related stories
arXiv AI
Sep 7
Forgetting Without Restarting: Execution-State Unlearning for Stateful LLM Agents
arXiv:2609. 04875v1 Announce Type: cross Abstract: Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache.
arXiv AI
Sep 4
Learning What Not to Forget: Long-Horizon Agent Memory from a Few Kilobytes of Learning
The paper introduces LRE (Learned Relevance Eviction), a lightweight, CPU‑only, language‑model‑free scorer that learns which parts of an agent’s interaction history are task‑critical and preserves them verbatim. In experiments, LRE matches or surpasses baseline eviction policies on accuracy‑cost trade‑offs, recovers 93% of full‑history accuracy, reduces worst‑case prompt size by 52%, and outperforms dense and token‑pruning encoders in conversational memory while being 295–1569× smaller. The method also achieves superior budgeted answer quality on LoCoMo reading and can be trained annotation‑free, recovering 95% of supervised scorer performance.
arXiv AI
Aug 25
Repo2Skill-Evo: Repository Skills Go Stale in Silence
arXiv:2608.21964v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly operate over evolving software repositories, where success depends on repository-specific procedural kno...
Hugging Face Trending Papers
Jul 7
Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantial inference compute before the failure becomes observable. We show that failure is predictable early from the agent's internal representations: lightweight per-round probes on hidden activations anticipate eventual episode failure as early as the first interaction round, where scorers reading only the agent's observable behavior are barely better than chance.
arXiv Machine Learning
Sep 10
Can an AI Assistant Really Forget? Auditable Deletion from Addressable Memory
This paper introduces a deletion interface for a pretrained language model, measuring how effectively deleted records are removed from the model’s memory. By retrofitting a support‑vector memory gate into the global attention layers of a frozen Gemma 3, the authors show that deletions can be performed without altering weights and that the resulting state is close to a reference state that never stored the record. Experiments on 4B‑parameter models demonstrate low perplexity impact and strong evidence that deleted content is hard to recover, while larger or smaller models fail to achieve the same guarantees.
arXiv Machine Learning
Jun 25
Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory
arXiv:2606. 25115v1 Announce Type: new Abstract: On-device language-model agents improve by accumulating experience in retrieved memory rather than by updating weights.
arXiv AI
Jun 29
Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents
arXiv:2606. 27472v1 Announce Type: cross Abstract: Large language model (LLM) agents operate over long, multi-session interactions in which facts change: a user moves, a price updates, a plan is revised.
arXiv AI
Sep 25
When Can Agents Forget Their Reasoning? ICLR for Long-Horizon Agent Context Compression
Long‑horizon language model agents accumulate reasoning history, which inflates context length and inference cost. The paper introduces Interaction Aware Compression for Long Horizon Reasoning (ICLR), a training‑free online method that ranks and removes reasoning blocks based on frozen proxy entropy while preserving actions, tool calls, and observations. On 260 WorkBuddyBench tasks, ICLR raises average reward from 0.699 to 0.718 and cuts input, output, and cache read tokens by 25.5%, 14.4%, and 33.3% respectively, while analyses show that historical reasoning becomes replaceable once task‑relevant state is externalized.
Hugging Face Trending Papers
Aug 12
Towards a Formal Definition of Agent Memory: Basis, Span, Optimality, and the Sequential Memory Problem
Despite the wide deployment of memory in large-model agents, there is no unified formal account of what a memory is or when it is optimal. This paper takes a first step toward this account.
arXiv AI
Aug 10
Coupling Planning with Episodic Memory in LLM Agents for Software Issue Resolution
arXiv:2608. 06811v1 Announce Type: cross Abstract: Resolving a real software issue with a large language model (LLM) agent is a long repair episode, often tens to hundreds of steps spanning exploration, hypothesis, implementation, and verification.
arXiv AI
4d ago
StateTape: Action-Conditioned Evidence Lifecycle Modeling for Long-Horizon Coding Agents
StateTape introduces a new framework for long‑horizon coding agents that rewrites the agent’s context as the code repository changes, rather than letting the context grow with every observation. It models the repository as a symbol‑level code graph, using a tape to mark symbols altered by each write and a manager model to resolve stale records. The authors provide theoretical analysis, a new benchmark called TraceBench, and empirical results showing higher resolve rates across six agents and three edit‑heavy benchmarks with minimal computational overhead.