Image de couverture de livre

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

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Imanov, Olaf Yunus Laitinen
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Computation and Language 68T05, 68T07, 68T50 I.2.6; I.2.7
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Strategic Doctrine Language Models (sdLM): A Learning-System Framework for Doctrinal Consistency and Geopolitical Forecasting
par: Imanov, Olaf Yunus Laitinen, et autres
Publié: (2026)

Harnessing non-adversarial robustness in large language models
par: Zhou, Qinghua, et autres
Publié: (2026)

Sentiment Analysis of Lithuanian Online Reviews Using Large Language Models
par: Vileikytė, Brigita, et autres
Publié: (2024)

Latent Object Permanence: Topological Phase Transitions, Free-Energy Principles, and Renormalization Group Flows in Deep Transformer Manifolds
par: Alpay, Faruk, et autres
Publié: (2026)

Inference acceleration for large language models using "stairs" assisted greedy generation
par: Grigaliūnas, Domas, et autres
Publié: (2024)

NOTAI.AI: Explainable Detection of Machine-Generated Text via Curvature and Feature Attribution
par: Breneur, Oleksandr Marchenko, et autres
Publié: (2026)

Rethinking the Multilingual Reasoning Gap with Layer Swap
par: Lasbordes, Maxence, et autres
Publié: (2026)

AIPsy-Affect: A Keyword-Free Clinical Stimulus Battery for Mechanistic Interpretability of Emotion in Language Models
par: Keeman, Michael
Publié: (2026)

CLMN: Concept based Language Models via Neural Symbolic Reasoning
par: Yang, Yibo
Publié: (2025)

SafeAnchor: Preventing Cumulative Safety Erosion in Continual Domain Adaptation of Large Language Models
par: Guo, Dongxin, et autres
Publié: (2026)

mHC-SSM: Manifold-Constrained Hyper-Connections for State Space Language Models with Stream-Specialized Adapters
par: Mutlu, Abdulvahap, et autres
Publié: (2026)

XAutoLM: Efficient Fine-Tuning of Language Models via Meta-Learning and AutoML
par: Estevanell-Valladares, Ernesto L., et autres
Publié: (2025)

How much do LLMs learn from negative examples?
par: Hamdan, Shadi, et autres
Publié: (2025)

Fine-tuning of Large Language Models for Constituency Parsing Using a Sequence to Sequence Approach
par: Delgado, Francisco Jose Cortes, et autres
Publié: (2025)

Vis-CoT: A Human-in-the-Loop Framework for Interactive Visualization and Intervention in LLM Chain-of-Thought Reasoning
par: Pather, Kaviraj, et autres
Publié: (2025)

Product-of-Experts Training Reduces Dataset Artifacts in Natural Language Inference
par: Mathew, Aby Mammen
Publié: (2026)

Extracting Sentence Embeddings from Pretrained Transformer Models
par: Stankevičius, Lukas, et autres
Publié: (2024)

ReFactor GNNs: Revisiting Factorisation-based Models from a Message-Passing Perspective
par: Chen, Yihong, et autres
Publié: (2022)

HEFT: A Coarse-to-Fine Hierarchy for Enhancing the Efficiency and Accuracy of Language Model Reasoning
par: Hill, Brennen
Publié: (2025)

When Does Content-Based Routing Work? Representation Requirements for Selective Attention in Hybrid Sequence Models
par: Basu, Abhinaba
Publié: (2026)

Linguistic Collapse: Neural Collapse in (Large) Language Models
par: Wu, Robert, et autres
Publié: (2024)

Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models
par: Fu, Tianyu, et autres
Publié: (2025)

How Pruning Reshapes Features: Sparse Autoencoder Analysis of Weight-Pruned Language Models
par: Borobia, Hector, et autres
Publié: (2026)

No Memorization, No Detection: Output Distribution-Based Contamination Detection in Small Language Models
par: Sela, Omer
Publié: (2026)

Enhancing Ultra-Low-Bit Quantization of Large Language Models Through Saliency-Aware Partial Retraining
par: Cao, Deyu, et autres
Publié: (2025)

ProbeScale: Probing Analysis to Optimize Neural Scaling Laws for Efficient Small Language Model Inference
par: Das, Sourav
Publié: (2026)

MAcPNN: Mutual Assisted Learning on Data Streams with Temporal Dependence
par: Giannini, Federico, et autres
Publié: (2026)

Unpacking Hateful Memes: Presupposed Context and False Claims
par: Cai, Weibin, et autres
Publié: (2025)

From Noise to Diversity: Random Embedding Injection in LLM Reasoning
par: Kim, Heejun, et autres
Publié: (2026)

CogniLoad: A Synthetic Natural Language Reasoning Benchmark With Tunable Length, Intrinsic Difficulty, and Distractor Density
par: Kaiser, Daniel, et autres
Publié: (2025)

Transactional Attention: Semantic Sponsorship for KV-Cache Retention
par: Basu, Abhinaba
Publié: (2026)

lmfaoooo at SemEval-2026 Task 1: Humor Is an Audience. Preference Modeling for Constrained Humor Generation
par: Tikhonov, Alexey, et autres
Publié: (2026)

Mitigating Position-Shift Failures in Text-Based Modular Arithmetic via Position Curriculum and Template Diversity
par: Yudin, Nikolay
Publié: (2026)

Who's Asking? Investigating Bias Through the Lens of Disability Framed Queries in LLMs
par: Hari, Vishnu, et autres
Publié: (2025)

The Geometry of Persona: Disentangling Personality from Reasoning in Large Language Models
par: Wang, Zhixiang
Publié: (2025)

How Human-Like Are Large Language Models? A Register-Aware Linguistic Evaluation Framework
par: Nieth, Björn, et autres
Publié: (2026)

The Concept Allocation Zone: Tracking How Concepts Form Across Transformer Depth
par: Henry, James
Publié: (2026)

Continuous Latent Diffusion Language Model
par: Guo, Hongcan, et autres
Publié: (2026)

A Practical Guide to Streaming Continual Learning
par: Cossu, Andrea, et autres
Publié: (2026)

$δ$-STEAL: LLM Stealing Attack with Local Differential Privacy
par: Dang, Kieu, et autres
Publié: (2025)