AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

Any-SSR: How Recursive Least Squares Works in Continual Learning of Large Language Model

Kai Tong, Kang Pan, Xiao Zhang, Erli Meng, Run He, Yawen Cui, Nuoyan Guo, Huiping Zhuang

ICCV 2025 · Continual learning › Continual fine-tuning of LLMs · approach: Low-rank and model-merging adaptation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Any-SSR instead of the best exemplar-free baselineLlama-2-7B-Chat · TRACE Order1OP: 55.69 vs 47.63 (SeqFT)Improved
use Any-SSR instead of replay-based SEEKRLlama-2-7B-Chat · TRACEOP: 54.99 -> 55.69 (Order1), 54.69 -> 55.69 (Order2)Improved
add analytic router with isolated LoRA subspaces (ablation)Any-SSR · TRACEBWT: -0.00 BWT in both orders vs -2.61 for SEEKRImproved
use analytic (AL) router instead of BP-trained router (ablation)Any-SSR · TRACE routing accuracyaverage accuracy drop per phase: 21.7% average accuracy drop per phase for BP router; 100% on previous tasks for analyticImproved
use separate LoRA adapters with analytic router instead of a single LoRA (ablation)Any-SSR · TRACEper-task accuracyImproved
move feature layer from Lf=2 to Lf=4 (E=10000) (ablation)Any-SSR · TRACE Order1OP: 54.79 -> 55.69Improved
move feature layer from Lf=4 to Lf=6 (E=10000) (ablation)Any-SSR · TRACE Order1OP: 55.69 -> 53.21Hurt
increase expansion size E from 5000 to 10000 (Lf=4) (ablation)Any-SSR · TRACE Order1OP: 54.89 -> 55.69Improved
increase expansion size E from 10000 to 15000 (Lf=4) (ablation)Any-SSR · TRACE Order1OP: 55.69 -> 55.69No clear effect
use Any-SSR instead of SeqFT for general abilityLlama-2-7B-Chat · General abilityGA: 43.50 (-4.27) -> 46.51 (-1.26)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy