Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling
- URL: http://arxiv.org/abs/2406.12585v1
- Date: Tue, 18 Jun 2024 13:17:26 GMT
- Title: Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling
- Authors: Yao-Ching Yu, Chun-Chih Kuo, Ziqi Ye, Yu-Cheng Chang, Yueh-Se Li,
- Abstract summary: In this paper, we treat the Generation of each token by Large Language Model (LLM) as a Classification (GaC) for ensembling.
In experiments, we ensemble state-of-the-art LLMs on several benchmarks, including exams, mathematics and reasoning, and observe that our method breaks the existing community performance ceiling.
- Score: 3.873482175367558
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Ensembling multiple models has always been an effective approach to push the limits of existing performance and is widely used in classification tasks by simply averaging the classification probability vectors from multiple classifiers to achieve better accuracy. However, in the thriving open-source Large Language Model (LLM) community, ensembling methods are rare and typically limited to ensembling the full-text outputs of LLMs, such as selecting the best output using a ranker, which leads to underutilization of token-level probability information. In this paper, we treat the Generation of each token by LLMs as a Classification (GaC) for ensembling. This approach fully exploits the probability information at each generation step and better prevents LLMs from producing early incorrect tokens that lead to snowballing errors. In experiments, we ensemble state-of-the-art LLMs on several benchmarks, including exams, mathematics and reasoning, and observe that our method breaks the existing community performance ceiling. Furthermore, we observed that most of the tokens in the answer are simple and do not affect the correctness of the final answer. Therefore, we also experimented with ensembling only key tokens, and the results showed better performance with lower latency across benchmarks.
Related papers
- Decoding with Limited Teacher Supervision Requires Understanding When to Trust the Teacher [11.136112399898481]
How can sLLMs efficiently utilize the supervision of LLMs to improve their generative quality?
We develop an algorithm to effectively aggregate the sLLM and LLM predictions on initial tokens so that the generated tokens can more accurately condition the subsequent token generation by sLLM only.
arXiv Detail & Related papers (2024-06-26T01:16:12Z) - Uncertainty Aware Learning for Language Model Alignment [97.36361196793929]
We propose uncertainty-aware learning (UAL) to improve the model alignment of different task scenarios.
We implement UAL in a simple fashion -- adaptively setting the label smoothing value of training according to the uncertainty of individual samples.
Experiments on widely used benchmarks demonstrate that our UAL significantly and consistently outperforms standard supervised fine-tuning.
arXiv Detail & Related papers (2024-06-07T11:37:45Z) - LLMEmbed: Rethinking Lightweight LLM's Genuine Function in Text Classification [13.319594321038926]
We propose a simple and effective transfer learning strategy, namely LLMEmbed, to address this classical but challenging task.
We perform extensive experiments on publicly available datasets, and the results show that LLMEmbed achieves strong performance while enjoys low training overhead.
arXiv Detail & Related papers (2024-06-06T03:46:59Z) - Cycles of Thought: Measuring LLM Confidence through Stable Explanations [53.15438489398938]
Large language models (LLMs) can reach and even surpass human-level accuracy on a variety of benchmarks, but their overconfidence in incorrect responses is still a well-documented failure mode.
We propose a framework for measuring an LLM's uncertainty with respect to the distribution of generated explanations for an answer.
arXiv Detail & Related papers (2024-06-05T16:35:30Z) - Language Model Cascades: Token-level uncertainty and beyond [65.38515344964647]
Recent advances in language models (LMs) have led to significant improvements in quality on complex NLP tasks.
Cascading offers a simple strategy to achieve more favorable cost-quality tradeoffs.
We show that incorporating token-level uncertainty through learned post-hoc deferral rules can significantly outperform simple aggregation strategies.
arXiv Detail & Related papers (2024-04-15T21:02:48Z) - How to Prune Your Language Model: Recovering Accuracy on the "Sparsity
May Cry'' Benchmark [60.72725673114168]
We revisit the question of accurate BERT-pruning during fine-tuning on downstream datasets.
We propose a set of general guidelines for successful pruning, even on the challenging SMC benchmark.
arXiv Detail & Related papers (2023-12-21T03:11:30Z) - Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
We reformulate open-ended generation tasks into token-level prediction tasks.
We instruct an LLM to self-evaluate its answers.
We benchmark a range of scoring methods based on self-evaluation.
arXiv Detail & Related papers (2023-12-14T19:09:22Z) - Alleviating Over-smoothing for Unsupervised Sentence Representation [96.19497378628594]
We present a Simple method named Self-Contrastive Learning (SSCL) to alleviate this issue.
Our proposed method is quite simple and can be easily extended to various state-of-the-art models for performance boosting.
arXiv Detail & Related papers (2023-05-09T11:00:02Z) - Easy Learning from Label Proportions [17.71834385754893]
Easyllp is a flexible and simple-to-implement debiasing approach based on aggregate labels.
Our technique allows us to accurately estimate the expected loss of an arbitrary model at an individual level.
arXiv Detail & Related papers (2023-02-06T20:41:38Z)
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.