Overtrained Language Models Are Harder to Fine-Tune

• URL: http://arxiv.org/abs/2503.19206v2
• Date: Fri, 28 Mar 2025 02:10:05 GMT
• Title: Overtrained Language Models Are Harder to Fine-Tune
• Abstract summary: Large language models are pre-trained on ever-growing token budgets.<n>We show that extended pre-training can make models harder to fine-tune, leading to degraded final performance.
• Score: 64.44743256512237
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models are pre-trained on ever-growing token budgets under the assumption that better pre-training performance translates to improved downstream models. In this work, we challenge this assumption and show that extended pre-training can make models harder to fine-tune, leading to degraded final performance. We term this phenomenon catastrophic overtraining. For example, the instruction-tuned OLMo-1B model pre-trained on 3T tokens leads to over 2% worse performance on multiple standard LLM benchmarks than its 2.3T token counterpart. Through controlled experiments and theoretical analysis, we show that catastrophic overtraining arises from a systematic increase in the broad sensitivity of pre-trained parameters to modifications, including but not limited to fine-tuning. Our findings call for a critical reassessment of pre-training design that considers the downstream adaptability of the model.

Related papers

• PriFT: Prior-Support Guided Supervised Fine-Tuning [74.65198014829393]
  Supervised fine-tuning (SFT) is an efficient approach for downstream task adaptation.<n>SFT fits fixed demonstrations token by token, including targets poorly aligned with the model's pretrained distribution.<n>We propose PriFT, which derives token weights from a frozen pretrained reference to obtain a stable reweighting signal unaffected by fine-tuning.
  arXiv  Detail & Related papers  (2026-06-08T12:14:06Z)
• 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)
• Nanbeige4-3B Technical Report: Exploring the Frontier of Small Language Models [23.832817775138675]
  Nanbeige4-3B is a family of small-scale but high-performing language models.<n>Pretrained on 23T high-quality tokens and finetuned on over 30 million diverse instructions, we extend the boundary of the scaling law for small language models.
  arXiv  Detail & Related papers  (2025-12-06T03:36:27Z)
• From Acceleration to Saturation: Scaling Behavior of Bootstrapped Language Model Pretraining [2.569647910019739]
  We study the scaling behavior of bootstrapped pretraining and find that its scaling efficiency diminishes in a predictable manner.<n>Our findings provide practical insights for efficient language model training and raise important considerations for the reuse of overtrained models.
  arXiv  Detail & Related papers  (2025-10-08T00:59:33Z)
• Amuro and Char: Analyzing the Relationship between Pre-Training and Fine-Tuning of Large Language Models [17.288865972774587]
  We investigate the relationship between pre-training and fine-tuning by fine-tuning multiple intermediate pre-trained model checkpoints.<n>Our results on 18 datasets suggest that pre-training improves the model in a latent way that unveils after fine-tuning.
  arXiv  Detail & Related papers  (2024-08-13T06:28:43Z)
• The Fine Line: Navigating Large Language Model Pretraining with Down-streaming Capability Analysis [27.310894780313618]
  This paper undertakes a comprehensive comparison of model capabilities at various pretraining intermediate checkpoints. We confirm that specific downstream metrics exhibit similar training dynamics across models of different sizes. In addition to our core findings, we've reproduced Amber and OpenLLaMA, releasing their intermediate checkpoints.
  arXiv  Detail & Related papers  (2024-04-01T16:00:01Z)
• 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)
• Bi-Drop: Enhancing Fine-tuning Generalization via Synchronous sub-net Estimation and Optimization [58.90989478049686]
  Bi-Drop is a fine-tuning strategy that selectively updates model parameters using gradients from various sub-nets. Experiments on the GLUE benchmark demonstrate that Bi-Drop consistently outperforms previous fine-tuning methods.
  arXiv  Detail & Related papers  (2023-05-24T06:09:26Z)
• Training Trajectories of Language Models Across Scales [99.38721327771208]
  Scaling up language models has led to unprecedented performance gains. How do language models of different sizes learn during pre-training? Why do larger language models demonstrate more desirable behaviors?
  arXiv  Detail & Related papers  (2022-12-19T19:16:29Z)
• Same Pre-training Loss, Better Downstream: Implicit Bias Matters for Language Models [46.24479693469042]
  This paper shows that 1) pre-training loss cannot fully explain downstream performance and 2) flatness of the model is well-correlated with downstream performance where pre-training loss is not.
  arXiv  Detail & Related papers  (2022-10-25T17:45:36Z)
• Adversarial Self-Attention for Language Understanding [89.265747130584]
  This paper proposes textitAdversarial Self-Attention mechanism (ASA). ASA adversarially reconstructs the Transformer attentions and facilitates model training from contaminated model structures. For fine-tuning, ASA-empowered models consistently outweigh naive models by a large margin considering both generalization and robustness.
  arXiv  Detail & Related papers  (2022-06-25T09:18:10Z)
• Accelerating Training of Transformer-Based Language Models with Progressive Layer Dropping [24.547833264405355]
  The proposed method achieves a 24% time reduction on average per sample and allows the pre-training to be 2.5 times faster than the baseline. While being faster, our pre-trained models are equipped with strong knowledge transferability, achieving comparable and sometimes higher GLUE score than the baseline.
  arXiv  Detail & Related papers  (2020-10-26T06:50:07Z)
• Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuning [134.15174177472807]
  We introduce adversarial training into self-supervision, to provide general-purpose robust pre-trained models for the first time. We conduct extensive experiments to demonstrate that the proposed framework achieves large performance margins.
  arXiv  Detail & Related papers  (2020-03-28T18:28:33Z)

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.