arXiv Machine Learning

Frozen in Time: How a Physics Foundation Model Forgets Under Sim-to-Real Adaptation

The paper investigates catastrophic forgetting in the scientific foundation model Walrus-1.3B during a two‑stage sim‑to‑real adaptation. Finetuning first on simulated flow around a cylinder quickly degrades performance on the original pretraining data, with one‑step errors rising up to 35× after 20 epochs, and further adaptation to experimental data worsens this degradation. The authors show that replaying as little as 1% of the simulated data during real‑data finetuning largely restores simulation performance while still adapting effectively to real dynamics.

arXiv Machine Learning

Oct 22

Combating Catastrophic Forgetting in Continual Domain Adaptation via Knowledge Recasting

The paper introduces Skill Schema Transport (SST), a post‑training framework that mitigates catastrophic forgetting in continual domain adaptation by translating new domain knowledge into a model’s existing skill space. SST abstracts each example type into a reusable skill schema, preserving essential operations while removing domain‑specific surface forms, thereby turning knowledge updates into procedural reuse rather than trajectory memorization. Experiments on two 3B instruction‑tuned backbones show that SST consistently reduces forgetting and improves generalization, achieving up to a 5.75‑point performance gain and a dramatic drop in maximum forgetting from 42.41 to 7.10 points on the Qwen‑3B MATH stream.

By Zijie Liu, Zhen Tan, Charles Fleming, Tianlong Chen

llms fine-tuning

arXiv Machine Learning

Oct 2

Replay on Demand: An Emergent Curriculum for Balancing Adaptation and Forgetting in Continued Pretraining

arXiv:2609.40089v1 Announce Type: new Abstract: Continued pretraining enables language models to adapt to new domains and knowledge, but often at the cost of forgetting previously acquired capabiliti...

By Lukas Thede, Shengzhuang Chen, Stefan Winzeck, Matthias Bethge, Zeynep Akata, Jonathan Richard Schwarz

llms

arXiv Machine Learning

Aug 13

Drift and Dependence: Layer-wise Information-Theoretic Bounds for Replay-Based Continual Learning

arXiv:2608. 11690v1 Announce Type: new Abstract: Continual learning must absorb new tasks without erasing old ones, and replay---mixing a small buffer of past examples into current training---is among the most effective remedies for catastrophic forgetting.

By Tieliang Gong, Zhongbo Zhang, Wen Wen, Yong-Jin Liu

benchmarks safety

arXiv AI

Jun 12

ARROW: Augmented Replay for RObust World models

arXiv:2603. 11395v3 Announce Type: replace-cross Abstract: Continual reinforcement learning challenges agents to acquire new skills while retaining previously learned ones with the goal of improving performance in both past and future tasks.

By Abdulaziz Alyahya, Abdallah Al Siyabi, Markus R. Ernst, Luke Yang, Levin Kuhlmann, Gideon Kowadlo

agents reinforcement-learning

arXiv Computer Vision

Oct 15

Data-Free Continual Learning for World Action Model with Rectified Generative Replay

arXiv:2610.18038v2 Announce Type: replace-cross Abstract: World action models (WAMs) built on pretrained video generation models adapt quickly to new robotic tasks, but they forget previously learned...

By Kaipeng Fang, Hangxu Liu, Chaojian Shi, Yunsong Zhou, Hongzi Zhu, Minyi Guo, Jingkuan Song, Lianli Gao

agents diffusion robotics fine-tuning

arXiv Machine Learning

Jun 29

Foundation vs. Specialized Models: Evaluating Catastrophic Forgetting in Continual Time Series Forecasting

arXiv:2510. 00809v3 Announce Type: replace Abstract: While Time Series Foundation Models (TSFMs) excel in zero-shot tasks, their behavior under continual fine tuning is poorly understood.

By Nouha Karaouli, Denis Coquenet, Elisa Fromont, Martial Mermillod, Marina Reyboz

fine-tuning benchmarks

arXiv Machine Learning

Jun 9

Theoretical Foundations of Continual Learning via Drift-Plus-Penalty

arXiv:2606. 08452v1 Announce Type: new Abstract: In many real-world settings, data streams are nonstationary and arrive sequentially, requiring learning systems to adapt continuously without retraining from scratch.

By Nazreen Shah, Govinda Arya, Bharath B. N., Ranjitha Prasad

benchmarks

arXiv Machine Learning

Jun 16

Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning

arXiv:2601. 18699v2 Announce Type: replace Abstract: Sequential fine-tuning of Large Language Models (LLMs) adaptation to target tasks often triggers catastrophic forgetting, where the acquisition of novel target skills degrades ancestral capabilities.

By Gustav Olaf Yunus Laitinen-Fredriksson Lundstrom-Imanov

llms fine-tuning benchmarks safety

arXiv Machine Learning

Jun 5

Catastrophic Forgetting as Accessibility Collapse: A Three-Level Framework for Knowledge Persistence in Continual Learning

arXiv:2606. 06032v1 Announce Type: new Abstract: Catastrophic forgetting is commonly interpreted as the irreversible erasure of previously acquired knowledge during sequential learning.

By Ayushman Trivedi, Bhavika Melwani

arXiv Machine Learning

Aug 18

Metaplasticity as adaptive gradient preconditioning for incremental learning

arXiv:2608. 14634v1 Announce Type: new Abstract: Biological intelligence naturally prevents catastrophic forgetting through Complementary Learning Systems (CLS) theory, a macroscopic consolidation process driven at the local level by synaptic metaplasticity: the continuous, history-dependent neuromodulation of individual synapses.

By Isabelle Aguilar, Zayn Andre Zainal, Omid Kavehei

benchmarks

arXiv Computation and Language

Oct 8

A Deafening Silence: Catastrophic Forgetting Lives in the Output Embeddings of Tokens the Data Never Speaks

arXiv:2610.09835v1 Announce Type: new Abstract: Continual pre-training and fine-tuning in Large Language Models (LLMs) inevitably induce catastrophic forgetting, typically mitigated by replay using o...

By Jonghyun Han, Younghoon Song, Jongyoul Park

llms rag fine-tuning

arXiv Machine Learning

Jun 25

Erased, but Not Gone: Output Forgetting Is Not True Forgetting

arXiv:2606. 25001v1 Announce Type: new Abstract: Machine unlearning (MU) is commonly judged by output forgetting, such as low forget-set accuracy or reduced logit-level membership inference.

By Teresa Pui Yee Yong, Win Kent Ong, Chee Seng Chan

Related stories