LadderEdit is a method for memory‑efficient lifelong editing of large language models that compresses each LoRA adapter after acquisition. Each edit is initially stored as a low‑rank sketch and only promoted to higher rank if it fails to satisfy rewrite, generalization, and locality contracts on probe prompts. On benchmarks such as ZsRE, CounterFact, and WikiBigEdit, LadderEdit achieves the same performance as exact LoRA while using 5.2× less memory and handling up to 50,000 sequential edits.
arXiv:2609.26219v1 Announce Type: cross Abstract: Long-running LLM agent workflows often revise interior context spans while retaining long suffixes. Although suffix tokens remain unchanged, altered...
LowRankArena is a standardized evaluation platform for SVD‑based low‑rank compression of large language models, unifying task versions, compression budgets, comparison regimes, and inference measurements. It provides a reproducible pipeline with over 3 TiB of released compressed checkpoints, enabling consistent comparisons across methods. An audit of five representative SVD techniques using LowRankArena shows that prior reported gains are highly conditional, with performance leaders and tiers shifting across backbones and keep ratios, and that nominal low‑rank savings often yield limited end‑to‑end speedups.
The paper introduces GEdit, a geometry-aware framework for editing large language models that leverages a Low-Rank Plus Diagonal factor model to capture the low-rank structure of Transformer FFN hidden representations. GEdit applies low‑rank regularization to protect core knowledge and adaptive regularization to adjust constraints based on the difficulty of each edit, allowing harder edits to succeed while preserving easier ones. Experiments on GPT2‑XL, GPT‑J, and LLaMA‑3‑8B with up to 10,000 continuous updates demonstrate that GEdit outperforms state‑of‑the‑art baselines in both editing success and knowledge preservation.
arXiv:2610.08811v1 Announce Type: new Abstract: As context windows scale to tens or hundreds of thousands of tokens, KV cache compression has become essential for efficient LLM inference. Existing me...
arXiv:2511. 05852v4 Announce Type: replace-cross Abstract: Knowledge editing (KE) offers a lightweight alternative to retraining for updating large language models (LLMs).
arXiv:2607. 28887v1 Announce Type: cross Abstract: Large language models increasingly write and repair production code, yet evidence is mounting that their test-passing patches leave codebases harder to maintain.
The paper compares two output regimes for code-editing language models: direct generation, where the model outputs the entire modified file, and iterative diff-based generation, where the model emits a sequence of localized edits. Experiments on Flutter/Dart tasks show that direct generation consistently outperforms diff-based generation across metrics such as compilation success, token efficiency, and quality judgments. However, diff-based generation can be competitive for short, spatially localized edits, particularly in refactoring and error-handling tasks with few edit steps.
The paper introduces EditJumps, the first open implementation of the Edit Flows and EvoFlows framework for antibody lead optimization. EditJumps trains a single generalist antibody editor on 1.66 M homolog pairs to generate homolog‑like variants of a seed sequence, enabling zero‑shot editing of unseen leads without per‑family retraining. The authors demonstrate that both Edit Flows and EvoFlows operate by sequentially applying edits at learned rates in continuous time, and they highlight the importance of open code for reproducing generative biology results, noting that published evaluation metrics can be highly sensitive to reference sample size.
arXiv:2607. 28947v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to solve complex problems by searching over program space, offering a general paradigm for scientific problems that can be naturally represented and solved as programs.
arXiv:2606. 26783v1 Announce Type: new Abstract: Fang et al.
arXiv:2606. 17107v1 Announce Type: cross Abstract: Prefix caching reuses prefill only across an exactly shared prefix, so one changed field invalidates the entire downstream cache.
arXiv:2609.25054v1 Announce Type: new Abstract: For a long-horizon LLM agent, the memory question is not what was once recorded but what \emph{currently holds}. Most designs answer it only indirectly...