Author name cluster

Lili Yu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

11 papers

2 author rows

NeurIPS Conference 2025 Conference Paper

CAT: Content-Adaptive Image Tokenization

Junhong Shen
Kushal Tirumala
Michihiro Yasunaga
Ishan Misra
Luke Zettlemoyer
Lili Yu
Chunting Zhou

Most existing image tokenizers encode images into a fixed number of tokens or patches, overlooking the inherent variability in image complexity and introducing unnecessary computate overhead for simpler images. To address this, we propose Content-Adaptive Tokenizer (CAT), which dynamically adjusts representation capacity based on the image content and encodes simpler images into fewer tokens. We design (1) a caption-based evaluation system that leverages LLMs to predict content complexity and determine the optimal compression ratio for an image, and (2) a novel nested VAE architecture that performs variable-rate compression in a single model. Trained on images with varying complexity, CAT achieves an average of 15% reduction in rFID across seven detail-rich datasets containing text, humans, and complex textures. On natural image datasets like ImageNet and COCO, it reduces token usage by 18% while maintaining high-fidelity reconstructions. We further evaluate CAT on two downstream tasks. For image classification, CAT consistently improves top-1 accuracy across five datasets spanning diverse domains. For image generation, it boosts training throughput by 23% on ImageNet, leading to more efficient learning and improved FIDs over fixed-token baselines.

PDF Details

NeurIPS Conference 2025 Conference Paper

Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better

Danny Driess
Jost Springenberg
Brian Ichter
Lili Yu
Adrian Li-Bell
Karl Pertsch
Allen Ren
Homer Walke

Vision-language-action (VLA) models provide a powerful approach to training control policies for physical systems, such as robots, by combining end-to-end learning with transfer of semantic knowledge from web-scale vision-language model (VLM) training. However, the constraints of real-time control are often at odds with the design of VLMs: the most powerful VLMs have tens or hundreds of billions of parameters, presenting an obstacle to real-time inference, and operate on discrete tokens rather than the continuous-valued outputs that are required for controlling robots. To address this challenge, recent VLA models have used specialized modules for efficient continuous control, such as action experts or continuous output heads, which typically require adding new untrained parameters to the pretrained VLM backbone. While these modules improve real-time and control capabilities, it remains an open question whether they preserve or degrade the semantic knowledge contained in the pretrained VLM, and what effect they have on the VLA training dynamics. In this paper, we study this question in the context of VLAs that include a continuous diffusion or flow matching action expert, showing that naively including such experts significantly harms both training speed and knowledge transfer. We provide an extensive analysis of various design choices, their impact on performance and knowledge transfer, and propose a technique for insulating the VLM backbone during VLA training that mitigates this issue. Videos are available at https: //pi. website/research/knowledge_insulation and open-source model weights are available at https: //github. com/Physical-Intelligence/openpi.

PDF Details

NeurIPS Conference 2025 Conference Paper

LMFusion: Adapting Pretrained Language Models for Multimodal Generation

Weijia Shi
Xiaochuang Han
Chunting Zhou
Weixin Liang
Xi Lin
Luke Zettlemoyer
Lili Yu

We present LMFusion, a framework for empowering pretrained text-only large language models (LLMs) with multimodal generative capabilities, enabling them to understand and generate both text and images in arbitrary sequences. LMFusion leverages existing Llama-3's weights for processing texts autoregressively while introducing additional and parallel transformer modules for processing images with diffusion. During training, the data from each modality is routed to its dedicated modules: modality-specific feedforward layers, query-key-value projections, and normalization layers process each modality independently, while the shared self-attention layers allow interactions across text and image features. By freezing the text-specific modules and only training the image-specific modules, LMFusion preserves the language capabilities of text-only LLMs while developing strong visual understanding and generation abilities. Compared to methods that pretrain multimodal generative models from scratch, our experiments demonstrate that, LMFusion improves image understanding by 20% and image generation by 3. 6% using only 50% of the FLOPs while maintaining Llama-3's language capabilities. We also demonstrate that this framework can adapt existing vision-language models with multimodal generation ability. Overall, this framework not only leverages existing computational investments in text-only LLMs but also enables the parallel development of language and vision capabilities, presenting a promising direction for efficient multimodal model development.

PDF Details

TMLR Journal 2025 Journal Article

Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models

Weixin Liang
Lili Yu
Liang Luo
Srini Iyer
Ning Dong
Chunting Zhou
Gargi Ghosh
Mike Lewis

The development of large language models (LLMs) has expanded to multi-modal systems capable of processing text, images, and speech within a unified framework. Training these models demands significantly larger datasets and computational resources compared to text-only LLMs. To address the scaling challenges, we introduce Mixture-of-Transformers (MoT), a sparse multi-modal transformer architecture that significantly reduces pretraining computational costs. MoT decouples non-embedding parameters of the model by modality -- including feed-forward networks, attention matrices, and layer normalization -- enabling modality-specific processing with global self-attention over the full input sequence. We evaluate MoT across multiple settings and model scales. In the Chameleon 7B setting (autoregressive text-and-image generation), MoT matches the dense baseline's performance using only 55.8% of the FLOPs. When extended to include speech, MoT reaches speech performance comparable to the dense baseline with only 37.2% of the FLOPs. In the Transfusion setting, where text and image are trained with different objectives, a 7B MoT model matches the image modality performance of the dense baseline with one third of the FLOPs, and a 760M MoT model outperforms a 1.4B dense baseline across key image generation metrics. System profiling further highlights MoT's practical benefits, achieving dense baseline image quality in 47.2% of the wall-clock time and text quality in 75.6% of the wall-clock time (measured on AWS p4de.24xlarge instances with NVIDIA A100 GPUs).

PDF Details

ICLR Conference 2025 Conference Paper

Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model

Chunting Zhou
Lili Yu
Arun Babu
Kushal Tirumala
Michihiro Yasunaga
Leonid Shamis
Jacob Kahn
Xuezhe Ma

We introduce Transfusion, a recipe for training a multi-modal model over discrete and continuous data. Transfusion combines the language modeling loss function (next token prediction) with diffusion to train a single transformer over mixed-modality sequences. We pretrain multiple Transfusion models up to 7B parameters from scratch on a mixture of text and image data, establishing scaling laws with respect to a variety of uni- and cross-modal benchmarks. Our experiments show that Transfusion scales significantly better than quantizing images and training a language model over discrete image tokens. By introducing modality-specific encoding and decoding layers, we can further improve the performance of Transfusion models, and even compress each image to just 16 patches. We further demonstrate that scaling our Transfusion recipe to 7B parameters and 2T multi-modal tokens produces a model that can generate images and text on a par with similar scale diffusion models and language models, reaping the benefits of both worlds.

Details

ICLR Conference 2024 Conference Paper

Jointly Training Large Autoregressive Multimodal Models

Emanuele Aiello
Lili Yu
Yixin Nie
Armen Aghajanyan
Barlas Oguz

In recent years, advances in the large-scale pretraining of language and text-to-image models have revolutionized the field of machine learning. Yet, integrating these two modalities into a single, robust model capable of generating seamless multimodal outputs remains a significant challenge. To address this gap, we present the Joint Autoregressive Mixture (JAM) framework, a modular approach that systematically fuses existing text and image generation models. We also introduce a specialized, data-efficient instruction-tuning strategy, tailored for mixed-modal generation tasks. Our final instruct-tuned model demonstrates unparalleled performance in generating high-quality multimodal outputs and represents the first model explicitly designed for this purpose.

Details

NeurIPS Conference 2024 Conference Paper

Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length

Xuezhe Ma
Xiaomeng Yang
Wenhan Xiong
Beidi Chen
Lili Yu
Hao Zhang
Jonathan May
Luke Zettlemoyer

The quadratic complexity and weak length extrapolation of Transformers limits their ability to scale to long sequences, and while sub-quadratic solutions like linear attention and state space models exist, they empirically underperform Transformers in pretraining efficiency and downstream task accuracy. We introduce MEGALODON, an neural architecture for efficient sequence modeling with unlimited context length. MEGALODON inherits the architecture of MEGA (exponential moving average with gated attention), and further introduces multiple technical components to improve its capability and stability, including complex exponential moving average (CEMA), timestep normalization layer, normalized attention mechanism and pre-norm with two-hop residual configuration. In a controlled head-to-head comparison with LLAMA2, MEGALODON achieves better efficiency than Transformer in the scale of 7 billion parameters and 2 trillion training tokens. MEGALODON reaches a training loss of 1. 70, landing mid-way between LLAMA2-7B (1. 75) and LLAMA2-13B (1. 67). This result is robust throughout a wide range of benchmarks, where MEGALODON consistently outperforms Transformers across different tasks, domains, and modalities.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

LIMA: Less Is More for Alignment

Chunting Zhou
Pengfei Liu
Puxin Xu
Srinivasan Iyer
Jiao Sun
Yuning Mao
Xuezhe Ma
Avia Efrat

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.

PDF Details

NeurIPS Conference 2023 Conference Paper

MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers

Lili Yu
Daniel Simig
Colin Flaherty
Armen Aghajanyan
Luke Zettlemoyer
Mike Lewis

Autoregressive transformers are spectacular models for short sequences but scale poorly to long sequences such as high-resolution images, podcasts, code, or books. We proposed Megabyte, a multi-scale decoder architecture that enables end-to-end differentiable modeling of sequences of over one million bytes. Megabyte segments sequences into patches and uses a local submodel within patches and a global model between patches. This enables sub-quadratic self-attention, much larger feedforward layers for the same compute, and improved parallelism during decoding---unlocking better performance at reduced cost for both training and generation. Extensive experiments show that Megabyte allows byte-level models to perform competitively with subword models on long context language modeling, achieve state-of-the-art density estimation on ImageNet, and model audio from raw files. Together, these results establish the viability of tokenization-free autoregressive sequence modeling at scale.

PDF Details

ICML Conference 2023 Conference Paper

Scaling Laws for Generative Mixed-Modal Language Models

Armen Aghajanyan
Lili Yu
Alexis Conneau
Wei-Ning Hsu
Karen Hambardzumyan
Susan Zhang
Stephen Roller
Naman Goyal 0001

Generative language models define distributions over sequences of tokens that can represent essentially any combination of data modalities (e. g. , any permutation of image tokens from VQ-VAEs, speech tokens from HuBERT, BPE tokens for language or code, and so on). To better understand the scaling properties of such mixed-modal models, we conducted over 250 experiments using seven different modalities and model sizes ranging from 8 million to 30 billion, trained on 5-100 billion tokens. We report new mixed-modal scaling laws that unify the contributions of individual modalities and the interactions between them. Specifically, we explicitly model the optimal synergy and competition due to data and model size as an additive term to previous uni-modal scaling laws. We also find four empirical phenomena observed during the training, such as emergent coordinate-ascent style training that naturally alternates between modalities, guidelines for selecting critical hyper-parameters, and connections between mixed-modal competition and training stability. Finally, we test our scaling law by training a 30B speech-text model, which significantly outperforms the corresponding unimodal models. Overall, our research provides valuable insights into the design and training of mixed-modal generative models, an important new class of unified models that have unique distributional properties.

Details

AAAI Conference 2021 Conference Paper

Nutri-bullets: Summarizing Health Studies by Composing Segments

Darsh J Shah
Lili Yu
Tao Lei
Regina Barzilay

We introduce Nutri-bullets, a multi-document summarization task for health and nutrition. First, we present two datasets of food and health summaries from multiple scientific studies. Furthermore, we propose a novel extract-compose model to solve the problem in the regime of limited parallel data. We explicitly select key spans from several abstracts using a policy network, followed by composing the selected spans to present a summary via a task specific language model. Compared to state-of-the-art methods, our approach leads to more faithful, relevant and diverse summarization – properties imperative to this application. For instance, on the BreastCancer dataset our approach gets a more than 50% improvement on relevance and faithfulness. 1

PDF Details