PALoRA: Projection-Adaptive LoRA for Preserving Reasoning in Large Language Models
The paper introduces PALoRA, a framework designed to enhance the integration of new knowledge into Large Language Models while preserving their reasoning capabilities. It addresses the challenge of maintaining reasoning performance amidst factual updates by employing a two-stage process. The results demonstrate that PALoRA retains a high level of reasoning performance while ensuring competitive factual recall across various benchmarks.
- ▪PALoRA stands for Projection-Adaptive LoRA and aims to tackle the plasticity-stability dilemma in Large Language Models.
- ▪The framework utilizes a Singular Value Fine-Tuning expert to identify critical components for reasoning before injecting new factual knowledge.
- ▪PALoRA has been shown to preserve 95% of reasoning performance while adding minimal parameter overhead.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.24549 |
| Publication time | Tue, 26 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-26T04:07:43.013Z |
| Last seen | 2026-05-26T04:07:43.013Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | ALX4vbGjpASb |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.24549 (cs) [Submitted on 23 May 2026] Title:PALoRA: Projection-Adaptive LoRA for Preserving Reasoning in Large Language Models Authors:Mustafa Hayri Bilgin, Mariam Barry, Albert Bifet, Azzedine Idir Ait Said, Soumya Banerjee View a PDF of the paper titled PALoRA: Projection-Adaptive LoRA for Preserving Reasoning in Large Language Models, by Mustafa Hayri Bilgin and 4 other authors View PDF HTML (experimental) Abstract:Efficiently updating Large Language Models (LLMs) with new or evolving factual knowledge remains a central challenge, as even parameter-efficient adaptation can erode previously acquired reasoning abilities.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
