CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training

• URL: http://arxiv.org/abs/2603.06610v1
• Date: Thu, 19 Feb 2026 09:46:24 GMT
• Title: CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training
• Abstract summary: We introduce textbfCapTrack, a capability-centric framework for analyzing forgetting in large language models.<n>We conduct a large-scale empirical study across post-training algorithms, domains, and model families.<n>We find that forgetting extends beyond parametric knowledge, with pronounced drift in robustness and default behaviors.
• Score: 48.70704477452434
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language model (LLM) post-training enhances latent skills, unlocks value alignment, improves performance, and enables domain adaptation. Unfortunately, post-training is known to induce forgetting, especially in the ubiquitous use-case of leveraging third-party pre-trained models, which is typically understood as a loss of parametric or factual knowledge. We argue that this accuracy-centric view is insufficient for modern foundation models and instead define forgetting as systematic model drift that degrades behavior and user experience. In this context, we introduce \textbf{CapTrack}, a capability-centric framework for analyzing forgetting in LLMs that combines a behavioral taxonomy with an evaluation suite built on established benchmarks and targeted adaptations. Using CapTrack, we conduct a large-scale empirical study across post-training algorithms, domains, and model families, including models up to 80B parameters. We find that forgetting extends beyond parametric knowledge, with pronounced drift in robustness and default behaviors. Instruction fine-tuning induces the strongest relative drift, while preference optimization is more conservative and can partially recover lost capabilities. Differences across model families persist, and no universal mitigation emerges.

Related papers

• D-CLING: Prior-Preserving Depth-Conditioned Fine-Tuning for Navigation Foundation Models [0.9831489366502301]
  Navigation Foundation Models (NFMs) trained on large cross-embodied datasets have demonstrated powerful generalizability in various scenarios.<n>Adopting in-domain fine-tuning for an NFM efficiently calibrates the visuomotor policy, promising further improvement even in a novel scenario.<n>We present a novel fine-tuning method that leverages large-scale pre-training while efficiently learning in novel setups.
  arXiv  Detail & Related papers  (2026-05-19T11:23:04Z)
• On Predicting the Post-training Potential of Pre-trained LLMs [60.07459271263409]
  We introduce a new task of predicting post-training potential - forecasting a base model's performance before post-training.<n>We propose RuDE, a unified framework that bypasses the generation gap of base models by leveraging response discrimination.<n>Experiments demonstrate a correlation greater than 90% with post-training performance.
  arXiv  Detail & Related papers  (2026-05-12T11:33:49Z)
• Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data [89.96277093034547]
  We introduce EntroDrop, an entropy-guided token dropout method that functions as structured data regularization.<n>We show that EntroDrop consistently outperforms standard regularization baselines and maintains robust performance throughout extended multi-epoch training.
  arXiv  Detail & Related papers  (2025-12-29T12:35:51Z)
• Parameter Importance-Driven Continual Learning for Foundation Models [5.471848114633189]
  Domain-specific post-training often causes catastrophic forgetting, making foundation models lose their general reasoning ability.<n>We introduce PIECE, a Importance Estimation-based Continual Enhancement method that preserves general ability while efficiently learning domain knowledge.<n>Our results highlight a practical path to scalable, domain-adaptive foundation models without catastrophic forgetting.
  arXiv  Detail & Related papers  (2025-11-19T12:07:53Z)
• Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual Learning [19.27175827358111]
  Continual learning in large language models (LLMs) is prone to catastrophic forgetting, where adapting to new tasks significantly degrades performance on previously learned ones.<n>We propose a novel continual full fine-tuning approach leveraging adaptive singular value decomposition (SVD)<n>We evaluate our approach extensively on standard continual learning benchmarks using both encoder-decoder (T5-Large) and decoder-only (LLaMA-2 7B) models.
  arXiv  Detail & Related papers  (2025-04-09T17:59:42Z)
• What Do Learning Dynamics Reveal About Generalization in LLM Reasoning? [83.83230167222852]
  We find that a model's generalization behavior can be effectively characterized by a training metric we call pre-memorization train accuracy. By connecting a model's learning behavior to its generalization, pre-memorization train accuracy can guide targeted improvements to training strategies.
  arXiv  Detail & Related papers  (2024-11-12T09:52:40Z)
• SMILE: Zero-Shot Sparse Mixture of Low-Rank Experts Construction From Pre-Trained Foundation Models [85.67096251281191]
  We present an innovative approach to model fusion called zero-shot Sparse MIxture of Low-rank Experts (SMILE) construction. SMILE allows for the upscaling of source models into an MoE model without extra data or further training. We conduct extensive experiments across diverse scenarios, such as image classification and text generation tasks, using full fine-tuning and LoRA fine-tuning.
  arXiv  Detail & Related papers  (2024-08-19T17:32:15Z)
• Information Guided Regularization for Fine-tuning Language Models [11.831883526217942]
  We argue that a more surgical approach to regularization needs to exist for smoother transfer learning. We devise a novel approach to dropout for improved model regularization and better downstream generalization.
  arXiv  Detail & Related papers  (2024-06-20T05:18:37Z)
• An Emulator for Fine-Tuning Large Language Models using Small Language Models [91.02498576056057]
  We introduce emulated fine-tuning (EFT), a principled and practical method for sampling from a distribution that approximates the result of pre-training and fine-tuning at different scales. We show that EFT enables test-time adjustment of competing behavioral traits like helpfulness and harmlessness without additional training. Finally, a special case of emulated fine-tuning, which we call LM up-scaling, avoids resource-intensive fine-tuning of large pre-trained models by ensembling them with small fine-tuned models.
  arXiv  Detail & Related papers  (2023-10-19T17:57:16Z)
• Value Gradient weighted Model-Based Reinforcement Learning [28.366157882991565]
  Model-based reinforcement learning (MBRL) is a sample efficient technique to obtain control policies. VaGraM is a novel method for value-aware model learning.
  arXiv  Detail & Related papers  (2022-04-04T13:28:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.