9/ StarCoder – an open-access 15.5B parameter LLM with 8K context length and is trained on large amounts of code spanning 80+ programming languages.
@dair_ai
-
FrugalGPT: Reducing LLM Inference Costs While Improving Performance
By
–
8/ FrugalGPT – presents strategies to reduce the inference cost associated with using LLMs while improving performance.
-
Active Retrieval Augmented LLMs Advance Knowledge-Intensive Generation
By
–
7/ Active Retrieval Augmented LLMs – an approach that actively decides when and what to retrieve across the course of the LLM generation; demonstrates superior or competitive performance on long-form knowledge-intensive generation tasks.
-
InstructBLIP Achieves State-of-the-Art Zero-Shot Visual-Language Performance
By
–
6/ InstructBLIP – explores visual-language instruction tuning based on the pre-trained BLIP-2 models; achieves state-of-the-art zero-shot performance on 13 held-out datasets, outperforming BLIP-2 and Flamingo.
-
TidyBot: Robot Learning User Preferences with LLM Planning
By
–
4/ TidyBot – shows that robots can combine language-based planning and perception with the few-shot summarization capabilities of LLMs to infer generalized user preferences that are applicable to future interactions.https://t.co/2Hv9FZh4Rx
— DAIR.AI (@dair_ai) 14 mai 20234/ TidyBot – shows that robots can combine language-based planning and perception with the few-shot summarization capabilities of LLMs to infer generalized user preferences that are applicable to future interactions.
-
Unfaithful Explanations in Chain-of-Thought Prompting
By
–
5/ Unfaithful Explanations in Chain-of-Thought Prompting – demonstrates that CoT explanations can misrepresent the true reason for a model’s prediction.
-
ImageBind: Joint Embedding Across Six Modalities
By
–
3/ ImageBind – an approach that learns a joint embedding data across 6 modalities at once; extends zero-shot capabilities to new modalities and enables emergent applications including cross-modal retrieval, composing modalities, and more.
-
Follow @dair_ai for top ML papers weekly
By
–
Follow @dair_ai to get the list of top ML papers every week.
-
Stable Low-Precision Training for Vision-Language Models
By
–
10/ Stable and Low-Precision Training for Large-Scale Vision-Language Models – introduces methods for accelerating and stabilizing training of large-scale language vision models.
-
ChatGPT Outperforms Physicians in Patient Response Quality
By
–
9/ Comparing Physicians vs ChatGPT – investigates if ChatGPT can provide quality responses to patient questions; finds that chatbot responses were preferred over physician responses and rated higher in terms of both quality and empathy.