Capability Salience Vector: Fine-grained Alignment of Loss and Capabilities for Downstream Task Scaling Law (2025.acl-long)
Copied to clipboard
Qiming Ge, Shuhao Xing, Songyang Gao, Yunhua Zhou, Yicheng Zou, Songyang Zhang, Zhi Chen, Hang Yan, Qi Zhang, Qipeng Guo, Kai Chen
| Challenge: | Large language models have demonstrated impressive performance across a wide range of tasks, but this achievement comes with the trade-off of significant computational demands. |
| Approach: | They propose a scaling law that decomposes the overall validation loss and assigns different importance weights to tokens to assess a specific meta-capability. |
| Outcome: | The proposed model can predict the loss trending of models across different levels of computation without a gap between validation loss and model's downstream capabilities. |
Similar Papers
Scaling Laws Are Unreliable for Downstream Tasks: A Reality Check (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Downstream scaling laws aim to predict task performance at larger scales from the model’s performance at smaller scales. |
| Approach: | They conduct a meta-analysis of existing data on downstream scaling laws and find that predictable scaling only occurs in a minority of cases: 39% of the time. |
| Outcome: | The proposed scaling laws only occur in a minority of cases, and seemingly benign changes to the experimental setting can completely change the scaling behavior. |
Not-Just-Scaling Laws: Towards a Better Understanding of the Downstream Impact of Language Model Design Decisions (2025.emnlp-main)
Copied to clipboard
Emmy Liu, Amanda Bertsch, Lintang Sutawika, Lindia Tjuatja, Patrick Fernandes, Lara Marinov, Michael Chen, Shreya Singhal, Carolin Lawrence, Aditi Raghunathan, Kiril Gashteovski, Graham Neubig
| Challenge: | Language model performance is largely dependent on pretraining decisions, but scaling laws based on only these two aspects do not always explain downstream task performance. |
| Approach: | They meta-analyze 92 open-source pretrained models to quantify their impact on performance. |
| Outcome: | The framework lays a foundation for more systematic investigation of how model development choices shape final capabilities. |
Revisiting Scaling Laws for Language Models: The Role of Data Quality and Training Strategies (2025.acl-long)
Copied to clipboard
| Challenge: | Existing scaling laws suggest augmenting model size and training data results in enhanced performance, but recent studies reveal deviations, particularly in large language models, where performance improvements decelerate—a phenomenon known as sub-scaling. |
| Approach: | They propose a sub-optimal scaling law that better predicts performance in sub-scaling regimes by examining data quality and training strategies. |
| Outcome: | The proposed scaling law better predicts performance in sub-scaling regimes, highlighting the importance of data quality and diversity. |
MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing generative language models neglect an inherent challenge in text corpus during training, i.e., the imbalance between frequent tokens and infrequent ones. |
| Approach: | They propose a function to mitigate the imbalance between frequent and infrequent tokens . authors propose 'MiLe Loss' function to assess learning difficulty of tokens during training . |
| Outcome: | Experiments show that models with proposed model can improve on downstream benchmarks. |
Collaborative Performance Prediction for Large Language Models (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are one of the most important AI research powered by largescale parameters, high computational resources, and massive training data. |
| Approach: | They propose a framework that leverages historical performance of large language models and other design factors to improve prediction accuracy. |
| Outcome: | The proposed framework surpasses scaling laws in predicting performance of large language models . it also facilitates a detailed analysis of factor importance, an area previously overlooked . |
Scaling Parameter-Constrained Language Models with Quality Data (2024.emnlp-industry)
Copied to clipboard
Ernie Chang, Matteo Paltenghi, Yang Li, Pin-Jie Lin, Changsheng Zhao, Patrick Huber, Zechun Liu, Rastislav Rabatin, Yangyang Shi, Vikas Chandra
| Challenge: | Scaling laws in language modeling quantify training loss as a function of dataset size and model parameters, but neglect the critical role of data quality in model generalization. |
| Approach: | They propose to use effective training tokens as a combination of text diversity and syntheticity as measured by a teacher model to calculate scaling laws. |
| Outcome: | The proposed term effective training tokens is a combination of two readily-computed indicators of text diversity and syntheticity as measured by a teacher model. |
Mitigating Catastrophic Forgetting in Large Language Models with Forgetting-aware Pruning (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) have shown impressive capabilities in various downstream tasks but typically face Catastrophic Forgetting (CF) during fine-tuning. |
| Approach: | They propose a pruning-based approach to balance CF and downstream task performance by integrating the ratio of the task vector to pre-trained model parameters into the pruning criteria. |
| Outcome: | The proposed pruning-based approach limits CF to just 0.25% while maintaining 99.67% accuracy on downstream tasks. |
Scaling Laws Under the Microscope: Predicting Transformer Performance from Small Scale Experiments (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Neural scaling laws define a predictable relationship between a model’s parameter count and its performance after training in the form of a power law. |
| Approach: | They perform an empirical investigation of language understanding tasks and evaluate their results to determine whether scaling laws can be used to accelerate model development. |
| Outcome: | The proposed scaling laws can be exploited for debugging convergence when training large models, and can predict the performance of larger models. |
How to inject knowledge efficiently? Knowledge Infusion Scaling Law for Pre-training Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies show that strategically infusing domain knowledge during pretraining can substantially improve downstream performance. |
| Approach: | They propose a knowledge infusion scaling law that predicts the optimal amount of domain knowledge to inject into large LLMs by analyzing their smaller counterparts. |
| Outcome: | The proposed model predicts the optimal amount of domain knowledge to inject into large LLMs by analyzing their smaller counterparts. |
Temporal Scaling Law for Large Language Models (2025.emnlp-main)
Copied to clipboard
Yizhe Xiong, Xiansheng Chen, Xin Ye, Hui Chen, Zijia Lin, Haoran Lian, Zhenpeng Su, Wei Huang, Jianwei Niu, Jungong Han, Guiguang Ding
| Challenge: | Existing studies have found that the test loss of LLMs scales as power-laws with model size, computational budget, and dataset size. |
| Approach: | They propose a concept of Temporal Scaling Law to study test loss of LLMs . they break down test loss into fine-grained token positions and develop a dynamic hyperbolic-law . |
| Outcome: | The proposed model predicts the test loss of LLMs as the training steps scale up. |