Mitigating Hallucination in LLMs This paper summarizes 32 techniques to mitigate hallucination in LLMs. Introduces a taxonomy categorizing methods like RAG, Knowledge Retrieval, CoVe, and more. Provides tips on how to apply these methods and highlights the challenges and

Excellent thoughts from Nabeel S. Qureshi on making the most of Twitter (which is still one of the very best things about the internet) nabeelqu.co/twitter

Like Andrew Ng, I have observed a definite shift in the prevalent discourse about AI at Davos: - Few people still talk about existential risk, and few people believe that current technology, even scaled up, will present an existential risk. - Everyone agrees that open source AI

Today we’re releasing Code Llama 70B: a new, more performant version of our LLM for code generation — available under the same license as previous Code Llama models. Download the models ➡️ bit.ly/3Oil6bQ • CodeLlama-70B • CodeLlama-70B-Python • CodeLlama-70B-Instruct

I am recruiting a PhD intern to work with me AI at Meta (GenAI) around LLMs, tool use, and / or the GAIA benchmark. Ideal profile: near the end of your PhD, willing to be based in Paris, starting in May/June 2024. DM if you’re interested!

Introducing Meta Llama 3: the most capable openly available LLM to date. Today we’re releasing 8B & 70B models that deliver on new capabilities such as improved reasoning and set a new state-of-the-art for models of their sizes. Today's release includes the first two Llama 3

🚨 Are We Done with MMLU? In our new paper "Are We Done with MMLU?" we identify errors in MMLU and find that some subsets are riddled with errors. We propose MMLU-Redux with 3,000 re-annotated questions across 30 subjects. 📜 arxiv.org/abs/2406.04127 (1/n)

I’m excited to share a few things we’re releasing today at Meta FAIR. These new AI model and dataset releases are part of our longstanding commitment to open science and I look forward to sharing even more work like this from the brilliant minds at FAIR! ai.meta.com/blog/meta-fair…

The team worked really hard to make history, voila finally the Llama-3.1 herd of models...have fun with it! * open 405B, insane 70B * 128K context length, improved reasoning & coding capabilities * detailed paper ai.meta.com/research/publi…

The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 👩‍🔬 arxiv.org/abs/2408.06292 It’s common for AI researchers to joke amongst themselves that “now all we need to do is figure out how to make AI write the papers for us!” but I think we’re now getting there!

Congrats to our team for winning two paper awards at #ACL2024! OLMo won the Best Theme Paper award, and Dolma won a Best Resource Paper award! All the credit goes to the whole team for the massive group effort 🎉🎉

Proud to release o1-preview to the world. Now that we have started to crack the challenge of getting models to “think” we are able to get large improvements on complex tasks by just letting them think harder.

Reinforcement learning with execution feedback (RLEF). Lots of sweat went into this one, but what works in principle works in practice: for code generation we can turn compute into training data: arxiv.org/abs/2410.02089 This works for LLMs, but will lead to world models.

(🎥45m) The Hitchhiker's Guide to Reasoning A talk about LLM reasoning, covering various methods, core problems, and future research directions! Covering topics such as: 🔍 The missing intermediates problem: how pre-training models are hobbled by missing reasoning steps, and

Super excited to share 🧠MLGym 🦾 – the first Gym environment for AI Research Agents 🤖🔬 We introduce MLGym and MLGym-Bench, a new framework and benchmark for evaluating and developing LLM agents on AI research tasks. The key contributions of our work are: 🕹️ Enables the

🎉 Excited to release General Reasoning: a new community resource for building open reasoning models. We’re looking to make personal, open reasoners a reality. Starting with a small step in that direction today! Read the thread in the quote tweet for details, or my personal

Today is the start of a new era of natively multimodal AI innovation. Today, we’re introducing the first Llama 4 models: Llama 4 Scout and Llama 4 Maverick — our most advanced models yet and the best in their class for multimodality. Llama 4 Scout • 17B-active-parameter model

Very excited that Aurora has been published in Nature: nature.com/articles/s4158…. Paper, code and model weights are all open and ready to be built on🚀

I’m building a new team at Google DeepMind to work on Open-Ended Discovery! We’re looking for strong Research Scientists and Research Engineers to help us push the frontier of autonomously discovering novel artifacts such as new knowledge, capabilities, or algorithms, in an

BREAKING: we've partnered with AI at Meta and Papers with Code to build a successor to Papers with Code (which was sunsetted yesterday) PWC, founded by Robert Stojnic and Ross Taylor has been an invaluable resource for AI scientists and engineers over the years (and an inspiration