![]() ![]() GPT-4, however, was still better than SAIL-7B. The models were fed web pages from Wikipedia and search results from DuckDuckGo to help them pick the right answers from a list of candidate responses. The experiments assessed their abilities to answer common sense and open-ended questions, as well as fact checking, and detecting hate speech. Initial experiments showed that SAIL-7B outperformed GPT-3.5 and other models containing more parameters at a range of tasks. This process filters out most unreliable and unrelated search results and improves the average instruction-following performance." "Our training explicitly includes a step that clarifies if each search result is helpful or not, and the language model follows the selected helpful information. As a result, our model can better summarize valuable information and generate better answers for various search queries, even when search engines cannot handle them very well," Luo said. "Our model learns to find helpful information from noisy search results and generate as accurate responses as possible. You can also play with a demo of the system hosted on Hugging Face. The details have been published in a paper released on arXiv, and the model's code is on GitHub. Luo said the fine-tuned model – nicknamed SAIL-7B, which stands for search-augmented instruction learning – is better at ignoring distracting or untrustworthy search results and generates higher quality answers. The researchers also constructed a separate dataset containing the top five web pages associated with each instruction, and trained the model to generate the correct response by ranking the sources on how relevant and closely aligned they were with the right response. The team tweaked Meta's LLaMA, a seven-billion-parameter LLM, fine-tuning it on a database containing 52,000 pairs of text-based instructions and corresponding responses generated by GPT-4. OpenAI calls for global watchdog focused on 'existential risk' posed by superintelligence.Red Hat promises AI trained on 'curated' and 'domain-specific' data.Texas judge demands lawyers declare AI-generated docs.Eating disorder non-profit pulls chatbot for emitting 'harmful advice'.Luo and his colleagues from MIT and the Chinese University of Hong Kong believe that models need to be fine-tuned further so they can better follow instructions on how to generate responses for web search. If Bard and Bing are to be useful, developers will need to figure out how to make LLMs extract the most useful information from a sea of text that is noisy, confusing and inconsistent.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |