New paper🚨 arxiv.org/abs/2212.08607 Can we build fluent, factual & logical text generation systems w/ multi-step reasoning over semi-structured data (tables/graphs)? We propose MURMUR, a neuro-symbolic method w/ symbolic modules for logical skills & LLMs for linguistic skills🧵

Introducing our new work, MURMUR, a neuro-symbolic modular reasoning approach to generate logical, faithful and diverse summaries from structured inputs w/ Swarnadeep Saha , Xinyan Velocity Yu , Mohit Bansal , Ramakanth

How do language models of different sizes learn during the course of pre-training? We study the training trajectories with training checkpoints of language model from 125M to 175B for a better understanding! Check out our new paper 📜: arxiv.org/abs/2212.09803 (1/N)

Announcing OPT-IML: a new language model from Meta AI with 175B parameters, fine-tuned on 2,000 language tasks — openly available soon under a noncommercial license for research use cases. Research paper & more details on GitHub ⬇️

Exciting updates on OPT-IML (arxiv.org/abs/2212.12017)! 1. The 175B models are now available to request for research purposes. Request here: (github.com/facebookresear…). 2. OPT-IML 30B and 1.3B models are now available on huggingface (huggingface.co/facebook/opt-i…)!

Our paper on 𝐩𝐫𝐨𝐦𝐩𝐭𝐢𝐧𝐠 𝐋𝐋𝐌𝐬 𝐰𝐢𝐭𝐡 𝐞𝐱𝐩𝐥𝐚𝐧𝐚𝐭𝐢𝐨𝐧𝐬 is accepted at Findings of #ACL2023NLP Check out our poster at the 𝐍𝐋𝐑𝐒𝐄 𝐰𝐨𝐫𝐤𝐬𝐡𝐨𝐩 𝐨𝐧 𝐓𝐡𝐮𝐫𝐬𝐝𝐚𝐲. I won't be there in person😢 but you can say hi to my advisor Greg Durrett 😆

I’m excited to release our most recent work setting a new SOTA FID of 4.88 on text-to-image generation we call CM3Leon (pronounced chameleon)! ai.meta.com/research/publi…

What if we combine PPO with Monte-Carlo Tree Search – the secret sauce for AlphaGo to reach superhuman performance? Spoiler: MAGIC!! Our inference-time decoding method, PPO-MCTS, achieves impressive results across many text generation tasks. 📜 arxiv.org/abs/2309.15028 🧵(1/n)

Long context models are popular, but is it the final solution to long text reading? We introduce a fundamentally different method, MemWalker: 1. Build a data structure (memory tree) 2. Traverse it via LLM prompting Outperforms long context, retrieval, & recurrent baselines. (1/n)

Introducing 🔮Crystal🔮, an LM that conducts “introspective reasoning” and shows its reasoning process for QA. This improves both QA accuracy and human interpretability => Reasoning made Crystal clear! arxiv.org/abs/2310.04921 Demo: hf.co/spaces/liujch1… at #EMNLP2023 🧵(1/n)

The ART of LLM Refinement: Ask, Refine, and Trust Achieves a performance gain of 5 points over self-refinement baselines, while using a much smaller model as the decision maker arxiv.org/abs/2311.07961

Newly published work from FAIR, Chameleon: Mixed-Modal Early-Fusion Foundation Models. This research presents a family of early-fusion token-based mixed-modal models capable of understanding & generating images & text in any arbitrary sequence. Paper ➡️ go.fb.me/7rb19n

Excited to release our work from last year showcasing a stable training recipe for fully token-based multi-modal early-fusion auto-regressive models! arxiv.org/abs/2405.09818 Huge shout out to Armen Aghajanyan Ramakanth Luke Zettlemoyer Gargi Ghosh and other co-authors. (1/n)

Today is a good day for open science. As part of our continued commitment to the growth and development of an open ecosystem, today at Meta FAIR we’re announcing four new publicly available AI models and additional research artifacts to inspire innovation in the community and

Super excited to open-source Chameleon 7B and 34B model weights today. These early-fusion models can understand and generate any sequence of interleaved images and text! Image-gen capabilities are masked for safety reasons, but everything else is enabled. Happy fine-tuning!

🚀 Introducing the Byte Latent Transformer (BLT) – An LLM architecture that scales better than Llama 3 using byte-patches instead of tokens 🤯 Paper 📄 dl.fbaipublicfiles.com/blt/BLT__Patch… Code 🛠️ github.com/facebookresear…

BLT model weights are out! Responding to popular demand, we just open-sourced model weights for our 1B and 8B BLT models for the research community to play with! huggingface.co/facebook/blt Hoping to see many new and improved BLT based architectures this year!

New research from FAIR- Active Reading: a framework to learn a given set of material with self-generated learning strategies for generalized and expert domains(such as Finance). Absorb significantly more knowledge than vanilla finetuning and usual data augmentations strategies

Meet HoneyBee: Our new 2.5M sample multi-modal reasoning dataset. It outperforms InternVL2.5/3-Instruct and Qwen2.5-VL-Instruct. More details in this post!

🧠 How can we equip LLMs with memory that allows them to continually learn new things? In our new paper with AI at Meta, we show how sparsely finetuning memory layers enables targeted updates for continual learning, w/ minimal interference with existing knowledge. While full