arXiv:2610.22381v1 Announce Type: cross Abstract: When a fact becomes harder to elicit from a language model, has the model lost it, or has it merely become less accessible? Parameter-updating test-time continual learning agents are usually evaluated by current output accuracy; w…
arXiv cs.AI
TIER_1English(EN)·Arnav Chavan, Kusha Sareen, Rishabh Tiwari, Abhranil Chandra, Johan Obando-Ceron, Kurt Keutzer, Scott Niekum, Inderjit S Dhillon, Devvrit Khatri·
arXiv:2610.22293v1 Announce Type: cross Abstract: Large language models (LLMs) acquire new skills through fine-tuning, but it can destroy capabilities they already have: catastrophic forgetting remains a central obstacle to continual learning in LLMs. Recent works connect forgett…
arXiv cs.LG
TIER_1English(EN)·Gilad Karpel, Daniel Soudry·
arXiv:2610.20804v1 Announce Type: new Abstract: We study continual linear regression in a high-dimensional regime, where a learner observes a sequence of tasks and updates its parameters sequentially. In this setting, we derive an explicit closed-form expression for the generaliz…
Large language models (LLMs) acquire new skills through fine-tuning, but it can destroy capabilities they already have: catastrophic forgetting remains a central obstacle to continual learning in LLMs. Recent works connect forgetting to how far, and along which directions, fine-t…
arXiv cs.LG
TIER_1English(EN)·Teif Aldaajani, Huda Hakami, Maha A. Thafar·
arXiv:2610.19876v1 Announce Type: new Abstract: Continual learning aims to acquire new knowledge while retaining previously learned knowledge. However, updating shared parameters when learning new tasks often overwrites existing knowledge, causing catastrophic forgetting. Memory …
arXiv:2506.03591v3 Announce Type: replace Abstract: Recent advances in unified autoregressive (AR) models have enabled multimodal understanding (MMU) and text-to-image generation (T2I) within a single architecture. However, in continual learning settings, their divergent objectiv…
