Decoding with Limited Teacher Supervision Requires Understanding When to Trust the Teacher
- URL: http://arxiv.org/abs/2406.18002v1
- Date: Wed, 26 Jun 2024 01:16:12 GMT
- Title: Decoding with Limited Teacher Supervision Requires Understanding When to Trust the Teacher
- Authors: Hyunjong Ok, Jegwang Ryu, Jaeho Lee,
- Abstract summary: 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.
- Score: 11.136112399898481
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: How can sLLMs efficiently utilize the supervision of LLMs to improve their generative quality? This question has been well studied in scenarios where there is no restriction on the number of LLM supervisions one can use, giving birth to many decoding algorithms that utilize supervision without further training. However, it is still unclear what is an effective strategy under the limited supervision scenario, where we assume that no more than a few tokens can be generated by LLMs. To this end, 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. Critically, we find that it is essential to adaptively overtrust or disregard the LLM prediction based on the confidence of the sLLM. Through our experiments on a wide range of models and datasets, we demonstrate that our method provides a consistent improvement over conventional decoding strategies.
Related papers
- Traditional Methods Outperform Generative LLMs at Forecasting Credit Ratings [17.109522466982476]
Large Language Models (LLMs) have been shown to perform well for many downstream tasks.
This paper investigates how well LLMs perform in the task of forecasting corporate credit ratings.
arXiv Detail & Related papers (2024-07-24T20:30:55Z) - Q*: Improving Multi-step Reasoning for LLMs with Deliberative Planning [53.6472920229013]
Large Language Models (LLMs) have demonstrated impressive capability in many natural language tasks.
LLMs are prone to produce errors, hallucinations and inconsistent statements when performing multi-step reasoning.
We introduce Q*, a framework for guiding LLMs decoding process with deliberative planning.
arXiv Detail & Related papers (2024-06-20T13:08:09Z) - One Token Can Help! Learning Scalable and Pluggable Virtual Tokens for Retrieval-Augmented Large Language Models [67.49462724595445]
Retrieval-augmented generation (RAG) is a promising way to improve large language models (LLMs)
We propose a novel method that involves learning scalable and pluggable virtual tokens for RAG.
arXiv Detail & Related papers (2024-05-30T03:44:54Z) - Tokenization Matters! Degrading Large Language Models through Challenging Their Tokenization [12.885866125783618]
Large Language Models (LLMs) tend to produce inaccurate responses to specific queries.
We construct an adversarial dataset, named as $textbfADT (Adrial dataset for Tokenizer)$ to challenge LLMs' tokenization.
Our empirical results reveal that our ADT is highly effective on challenging the tokenization of leading LLMs, including GPT-4o, Llama-3, Qwen2.5-max and so on.
arXiv Detail & Related papers (2024-05-27T11:39:59Z) - How Can LLM Guide RL? A Value-Based Approach [68.55316627400683]
Reinforcement learning (RL) has become the de facto standard practice for sequential decision-making problems by improving future acting policies with feedback.
Recent developments in large language models (LLMs) have showcased impressive capabilities in language understanding and generation, yet they fall short in exploration and self-improvement capabilities.
We develop an algorithm named LINVIT that incorporates LLM guidance as a regularization factor in value-based RL, leading to significant reductions in the amount of data needed for learning.
arXiv Detail & Related papers (2024-02-25T20:07:13Z) - TRACE: A Comprehensive Benchmark for Continual Learning in Large
Language Models [52.734140807634624]
Aligned large language models (LLMs) demonstrate exceptional capabilities in task-solving, following instructions, and ensuring safety.
Existing continual learning benchmarks lack sufficient challenge for leading aligned LLMs.
We introduce TRACE, a novel benchmark designed to evaluate continual learning in LLMs.
arXiv Detail & Related papers (2023-10-10T16:38:49Z) - LLMRec: Benchmarking Large Language Models on Recommendation Task [54.48899723591296]
The application of Large Language Models (LLMs) in the recommendation domain has not been thoroughly investigated.
We benchmark several popular off-the-shelf LLMs on five recommendation tasks, including rating prediction, sequential recommendation, direct recommendation, explanation generation, and review summarization.
The benchmark results indicate that LLMs displayed only moderate proficiency in accuracy-based tasks such as sequential and direct recommendation.
arXiv Detail & Related papers (2023-08-23T16:32:54Z) - On Learning to Summarize with Large Language Models as References [101.79795027550959]
Large language models (LLMs) are favored by human annotators over the original reference summaries in commonly used summarization datasets.
We study an LLM-as-reference learning setting for smaller text summarization models to investigate whether their performance can be substantially improved.
arXiv Detail & Related papers (2023-05-23T16:56:04Z) - Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM
Inference Pipeline [22.08897444328099]
Large language models (LLMs) have revolutionized the field of AI, demonstrating unprecedented capacity across various tasks.
In this paper, we propose an efficient LLM inference pipeline that harnesses the power of LLMs.
arXiv Detail & Related papers (2023-05-22T15:36:06Z)
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.