Alert button
Picture for Gargi Ghosh

Gargi Ghosh

Alert button

Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning

Sep 05, 2023
Lili Yu, Bowen Shi, Ramakanth Pasunuru, Benjamin Muller, Olga Golovneva, Tianlu Wang, Arun Babu, Binh Tang, Brian Karrer, Shelly Sheynin, Candace Ross, Adam Polyak, Russell Howes, Vasu Sharma, Puxin Xu, Hovhannes Tamoyan, Oron Ashual, Uriel Singer, Shang-Wen Li, Susan Zhang, Richard James, Gargi Ghosh, Yaniv Taigman, Maryam Fazel-Zarandi, Asli Celikyilmaz, Luke Zettlemoyer, Armen Aghajanyan

Figure 1 for Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning
Figure 2 for Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning
Figure 3 for Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning
Figure 4 for Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning

We present CM3Leon (pronounced "Chameleon"), a retrieval-augmented, token-based, decoder-only multi-modal language model capable of generating and infilling both text and images. CM3Leon uses the CM3 multi-modal architecture but additionally shows the extreme benefits of scaling up and tuning on more diverse instruction-style data. It is the first multi-modal model trained with a recipe adapted from text-only language models, including a large-scale retrieval-augmented pre-training stage and a second multi-task supervised fine-tuning (SFT) stage. It is also a general-purpose model that can do both text-to-image and image-to-text generation, allowing us to introduce self-contained contrastive decoding methods that produce high-quality outputs. Extensive experiments demonstrate that this recipe is highly effective for multi-modal models. CM3Leon achieves state-of-the-art performance in text-to-image generation with 5x less training compute than comparable methods (zero-shot MS-COCO FID of 4.88). After SFT, CM3Leon can also demonstrate unprecedented levels of controllability in tasks ranging from language-guided image editing to image-controlled generation and segmentation.

Viaarxiv icon

LIMA: Less Is More for Alignment

May 18, 2023
Chunting Zhou, Pengfei Liu, Puxin Xu, Srini Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, Susan Zhang, Gargi Ghosh, Mike Lewis, Luke Zettlemoyer, Omer Levy

Figure 1 for LIMA: Less Is More for Alignment
Figure 2 for LIMA: Less Is More for Alignment
Figure 3 for LIMA: Less Is More for Alignment
Figure 4 for LIMA: Less Is More for Alignment

Large language models are trained in two stages: (1) unsupervised pretraining from raw text, to learn general-purpose representations, and (2) large scale instruction tuning and reinforcement learning, to better align to end tasks and user preferences. We measure the relative importance of these two stages by training LIMA, a 65B parameter LLaMa language model fine-tuned with the standard supervised loss on only 1,000 carefully curated prompts and responses, without any reinforcement learning or human preference modeling. LIMA demonstrates remarkably strong performance, learning to follow specific response formats from only a handful of examples in the training data, including complex queries that range from planning trip itineraries to speculating about alternate history. Moreover, the model tends to generalize well to unseen tasks that did not appear in the training data. In a controlled human study, responses from LIMA are either equivalent or strictly preferred to GPT-4 in 43% of cases; this statistic is as high as 58% when compared to Bard and 65% versus DaVinci003, which was trained with human feedback. Taken together, these results strongly suggest that almost all knowledge in large language models is learned during pretraining, and only limited instruction tuning data is necessary to teach models to produce high quality output.

Viaarxiv icon

CiT: Curation in Training for Effective Vision-Language Data

Jan 05, 2023
Hu Xu, Saining Xie, Po-Yao Huang, Licheng Yu, Russell Howes, Gargi Ghosh, Luke Zettlemoyer, Christoph Feichtenhofer

Figure 1 for CiT: Curation in Training for Effective Vision-Language Data
Figure 2 for CiT: Curation in Training for Effective Vision-Language Data
Figure 3 for CiT: Curation in Training for Effective Vision-Language Data
Figure 4 for CiT: Curation in Training for Effective Vision-Language Data

Large vision-language models are generally applicable to many downstream tasks, but come at an exorbitant training cost that only large institutions can afford. This paper trades generality for efficiency and presents Curation in Training (CiT), a simple and efficient vision-text learning algorithm that couples a data objective into training. CiT automatically yields quality data to speed-up contrastive image-text training and alleviates the need for an offline data filtering pipeline, allowing broad data sources (including raw image-text pairs from the web). CiT contains two loops: an outer loop curating the training data and an inner loop consuming the curated training data. The text encoder connects the two loops. Given metadata for tasks of interest, e.g., class names, and a large pool of image-text pairs, CiT alternatively selects relevant training data from the pool by measuring the similarity of their text embeddings and embeddings of the metadata. In our experiments, we observe that CiT can speed up training by over an order of magnitude, especially if the raw data size is large.

* Technical Report 
Viaarxiv icon

ALERT: Adapting Language Models to Reasoning Tasks

Dec 16, 2022
Ping Yu, Tianlu Wang, Olga Golovneva, Badr Alkhamissy, Gargi Ghosh, Mona Diab, Asli Celikyilmaz

Figure 1 for ALERT: Adapting Language Models to Reasoning Tasks
Figure 2 for ALERT: Adapting Language Models to Reasoning Tasks
Figure 3 for ALERT: Adapting Language Models to Reasoning Tasks
Figure 4 for ALERT: Adapting Language Models to Reasoning Tasks

Current large language models can perform reasonably well on complex tasks that require step-by-step reasoning with few-shot learning. Are these models applying reasoning skills they have learnt during pre-training and reason outside of their training context, or are they simply memorizing their training corpus at finer granularity and have learnt to better understand their context? To tease apart these possibilities, we introduce ALERT, a benchmark and suite of analyses for assessing language models' reasoning ability comparing pre-trained and finetuned models on complex tasks that require reasoning skills to solve. ALERT provides a test bed to asses any language model on fine-grained reasoning skills, which spans over 20 datasets and covers 10 different reasoning skills. We leverage ALERT to further investigate the role of finetuning. With extensive empirical analysis we find that language models learn more reasoning skills such as textual entailment, abductive reasoning, and analogical reasoning during finetuning stage compared to pretraining state. We also find that when language models are finetuned they tend to overfit to the prompt template, which hurts the robustness of models causing generalization problems.

Viaarxiv icon

MAViL: Masked Audio-Video Learners

Dec 15, 2022
Po-Yao Huang, Vasu Sharma, Hu Xu, Chaitanya Ryali, Haoqi Fan, Yanghao Li, Shang-Wen Li, Gargi Ghosh, Jitendra Malik, Christoph Feichtenhofer

Figure 1 for MAViL: Masked Audio-Video Learners
Figure 2 for MAViL: Masked Audio-Video Learners
Figure 3 for MAViL: Masked Audio-Video Learners
Figure 4 for MAViL: Masked Audio-Video Learners

We present Masked Audio-Video Learners (MAViL) to train audio-visual representations. Our approach learns with three complementary forms of self-supervision: (1) reconstruction of masked audio and video input data, (2) intra- and inter-modal contrastive learning with masking, and (3) self-training by reconstructing joint audio-video contextualized features learned from the first two objectives. Pre-training with MAViL not only enables the model to perform well in audio-visual classification and retrieval tasks but also improves representations of each modality in isolation, without using information from the other modality for fine-tuning or inference. Empirically, MAViL sets a new state-of-the-art on AudioSet (53.1 mAP) and VGGSound (67.1% accuracy). For the first time, a self-supervised audio-visual model outperforms ones that use external supervision on these benchmarks. Code will be available soon.

* Technical report 
Viaarxiv icon

CM3: A Causal Masked Multimodal Model of the Internet

Jan 19, 2022
Armen Aghajanyan, Bernie Huang, Candace Ross, Vladimir Karpukhin, Hu Xu, Naman Goyal, Dmytro Okhonko, Mandar Joshi, Gargi Ghosh, Mike Lewis, Luke Zettlemoyer

Figure 1 for CM3: A Causal Masked Multimodal Model of the Internet
Figure 2 for CM3: A Causal Masked Multimodal Model of the Internet
Figure 3 for CM3: A Causal Masked Multimodal Model of the Internet
Figure 4 for CM3: A Causal Masked Multimodal Model of the Internet

We introduce CM3, a family of causally masked generative models trained over a large corpus of structured multi-modal documents that can contain both text and image tokens. Our new causally masked approach generates tokens left to right while also masking out a small number of long token spans that are generated at the end of the string, instead of their original positions. The casual masking object provides a type of hybrid of the more common causal and masked language models, by enabling full generative modeling while also providing bidirectional context when generating the masked spans. We train causally masked language-image models on large-scale web and Wikipedia articles, where each document contains all of the text, hypertext markup, hyperlinks, and image tokens (from a VQVAE-GAN), provided in the order they appear in the original HTML source (before masking). The resulting CM3 models can generate rich structured, multi-modal outputs while conditioning on arbitrary masked document contexts, and thereby implicitly learn a wide range of text, image, and cross modal tasks. They can be prompted to recover, in a zero-shot fashion, the functionality of models such as DALL-E, GENRE, and HTLM. We set the new state-of-the-art in zero-shot summarization, entity linking, and entity disambiguation while maintaining competitive performance in the fine-tuning setting. We can generate images unconditionally, conditioned on text (like DALL-E) and do captioning all in a zero-shot setting with a single model.

Viaarxiv icon

VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding

Oct 01, 2021
Hu Xu, Gargi Ghosh, Po-Yao Huang, Dmytro Okhonko, Armen Aghajanyan, Florian Metze, Luke Zettlemoyer, Christoph Feichtenhofer

Figure 1 for VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding
Figure 2 for VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding
Figure 3 for VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding
Figure 4 for VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding

We present VideoCLIP, a contrastive approach to pre-train a unified model for zero-shot video and text understanding, without using any labels on downstream tasks. VideoCLIP trains a transformer for video and text by contrasting temporally overlapping positive video-text pairs with hard negatives from nearest neighbor retrieval. Our experiments on a diverse series of downstream tasks, including sequence-level text-video retrieval, VideoQA, token-level action localization, and action segmentation reveal state-of-the-art performance, surpassing prior work, and in some cases even outperforming supervised approaches. Code is made available at https://github.com/pytorch/fairseq/tree/main/examples/MMPT.

* EMNLP 2021 
Viaarxiv icon

HTLM: Hyper-Text Pre-Training and Prompting of Language Models

Jul 14, 2021
Armen Aghajanyan, Dmytro Okhonko, Mike Lewis, Mandar Joshi, Hu Xu, Gargi Ghosh, Luke Zettlemoyer

Figure 1 for HTLM: Hyper-Text Pre-Training and Prompting of Language Models
Figure 2 for HTLM: Hyper-Text Pre-Training and Prompting of Language Models
Figure 3 for HTLM: Hyper-Text Pre-Training and Prompting of Language Models
Figure 4 for HTLM: Hyper-Text Pre-Training and Prompting of Language Models

We introduce HTLM, a hyper-text language model trained on a large-scale web crawl. Modeling hyper-text has a number of advantages: (1) it is easily gathered at scale, (2) it provides rich document-level and end-task-adjacent supervision (e.g. class and id attributes often encode document category information), and (3) it allows for new structured prompting that follows the established semantics of HTML (e.g. to do zero-shot summarization by infilling title tags for a webpage that contains the input text). We show that pretraining with a BART-style denoising loss directly on simplified HTML provides highly effective transfer for a wide range of end tasks and supervision levels. HTLM matches or exceeds the performance of comparably sized text-only LMs for zero-shot prompting and fine-tuning for classification benchmarks, while also setting new state-of-the-art performance levels for zero-shot summarization. We also find that hyper-text prompts provide more value to HTLM, in terms of data efficiency, than plain text prompts do for existing LMs, and that HTLM is highly effective at auto-prompting itself, by simply generating the most likely hyper-text formatting for any available training data. We will release all code and models to support future HTLM research.

Viaarxiv icon

VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding

May 20, 2021
Hu Xu, Gargi Ghosh, Po-Yao Huang, Prahal Arora, Masoumeh Aminzadeh, Christoph Feichtenhofer, Florian Metze, Luke Zettlemoyer

Figure 1 for VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding
Figure 2 for VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding
Figure 3 for VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding
Figure 4 for VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding

We present a simplified, task-agnostic multi-modal pre-training approach that can accept either video or text input, or both for a variety of end tasks. Existing pre-training are task-specific by adopting either a single cross-modal encoder that requires both modalities, limiting their use for retrieval-style end tasks or more complex multitask learning with two unimodal encoders, limiting early cross-modal fusion. We instead introduce new pretraining masking schemes that better mix across modalities (e.g. by forcing masks for text to predict the closest video embeddings) while also maintaining separability (e.g. unimodal predictions are sometimes required, without using all the input). Experimental results show strong performance across a wider range of tasks than any previous methods, often outperforming task-specific pre-training.

* 9 pages, ACL Findings 2021 
Viaarxiv icon

Multi-task Retrieval for Knowledge-Intensive Tasks

Jan 01, 2021
Jean Maillard, Vladimir Karpukhin, Fabio Petroni, Wen-tau Yih, Barlas Oğuz, Veselin Stoyanov, Gargi Ghosh

Figure 1 for Multi-task Retrieval for Knowledge-Intensive Tasks
Figure 2 for Multi-task Retrieval for Knowledge-Intensive Tasks
Figure 3 for Multi-task Retrieval for Knowledge-Intensive Tasks
Figure 4 for Multi-task Retrieval for Knowledge-Intensive Tasks

Retrieving relevant contexts from a large corpus is a crucial step for tasks such as open-domain question answering and fact checking. Although neural retrieval outperforms traditional methods like tf-idf and BM25, its performance degrades considerably when applied to out-of-domain data. Driven by the question of whether a neural retrieval model can be universal and perform robustly on a wide variety of problems, we propose a multi-task trained model. Our approach not only outperforms previous methods in the few-shot setting, but also rivals specialised neural retrievers, even when in-domain training data is abundant. With the help of our retriever, we improve existing models for downstream tasks and closely match or improve the state of the art on multiple benchmarks.

Viaarxiv icon