Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929600126255104 |
|---|---|
| author | Chen, Haolin Feng, Yihao Liu, Zuxin Yao, Weiran Prabhakar, Akshara Heinecke, Shelby Ho, Ricky Mui, Phil Savarese, Silvio Xiong, Caiming Wang, Huan |
| author_facet | Chen, Haolin Feng, Yihao Liu, Zuxin Yao, Weiran Prabhakar, Akshara Heinecke, Shelby Ho, Ricky Mui, Phil Savarese, Silvio Xiong, Caiming Wang, Huan |
| contents | Large language models (LLMs) have shown impressive capabilities, but still struggle with complex reasoning tasks requiring multiple steps. While prompt-based methods like Chain-of-Thought (CoT) can improve LLM reasoning at inference time, optimizing reasoning capabilities during training remains challenging. We introduce LaTent Reasoning Optimization (LaTRO), a principled framework that formulates reasoning as sampling from a latent distribution and optimizes it via variational approaches. LaTRO enables LLMs to concurrently improve both their reasoning process and ability to evaluate reasoning quality, without requiring external feedback or reward models. We validate LaTRO through experiments on GSM8K and ARC-Challenge datasets using multiple model architectures. On GSM8K, LaTRO improves zero-shot accuracy by an average of 12.5% over base models and 9.6% over supervised fine-tuning across Phi-3.5-mini, Mistral-7B, and Llama-3.1-8B. Our findings suggest that pre-trained LLMs possess latent reasoning capabilities that can be unlocked and enhanced through our proposed optimization approach in a self-improvement manner. The code of LaTRO is available at \url{https://github.com/SalesforceAIResearch/LaTRO}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_04282 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding Chen, Haolin Feng, Yihao Liu, Zuxin Yao, Weiran Prabhakar, Akshara Heinecke, Shelby Ho, Ricky Mui, Phil Savarese, Silvio Xiong, Caiming Wang, Huan Artificial Intelligence Computation and Language Machine Learning I.2.7 Large language models (LLMs) have shown impressive capabilities, but still struggle with complex reasoning tasks requiring multiple steps. While prompt-based methods like Chain-of-Thought (CoT) can improve LLM reasoning at inference time, optimizing reasoning capabilities during training remains challenging. We introduce LaTent Reasoning Optimization (LaTRO), a principled framework that formulates reasoning as sampling from a latent distribution and optimizes it via variational approaches. LaTRO enables LLMs to concurrently improve both their reasoning process and ability to evaluate reasoning quality, without requiring external feedback or reward models. We validate LaTRO through experiments on GSM8K and ARC-Challenge datasets using multiple model architectures. On GSM8K, LaTRO improves zero-shot accuracy by an average of 12.5% over base models and 9.6% over supervised fine-tuning across Phi-3.5-mini, Mistral-7B, and Llama-3.1-8B. Our findings suggest that pre-trained LLMs possess latent reasoning capabilities that can be unlocked and enhanced through our proposed optimization approach in a self-improvement manner. The code of LaTRO is available at \url{https://github.com/SalesforceAIResearch/LaTRO}. |
| title | Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding |
| topic | Artificial Intelligence Computation and Language Machine Learning I.2.7 |
| url | https://arxiv.org/abs/2411.04282 |