Sanjiv Kumar
Hi! I work in the area of large-scale machine learning and computer vision. You can find more information about me including a complete list of papers at: www.sanjivk.com.
Authored Publications
Sort By
Preview abstract
Large language models (LLMs) have shown strong results on a range of applications,
including regression and scoring tasks. Typically, one obtains outputs from an LLM via autoregressive sampling from the model’s output distribution. We show that this inference
strategy can be sub-optimal for common regression and scoring evaluation metrics. As a
remedy, we build on prior work on Minimum Bayes Risk decoding, and propose alternate
inference strategies that estimate the Bayes-optimal solution for regression and scoring metrics in closed-form from sampled responses. We show that our proposal significantly improves over baselines across datasets and models.
View details
DistillSpec: Improving speculative decoding via knowledge distillation
Yongchao Zhou
Kaifeng Lyu
Aditya Menon
Jean-François Kagy
International Conference on Learning Representations (ICLR) (2024)
Preview abstract
Speculative decoding proves highly effective in expediting Large Language Model inference by employing a smaller draft model for token generation and a larger model for parallel token verification. Nonetheless, identifying an accurate and compact draft model aligned with the target model presents challenges. To address this, we propose leveraging white-box knowledge distillation, significantly improving draft model alignment with the larger target model, thereby enhancing speculative decoding. Our findings underscore the pivotal role of on-policy data generation and a suitable divergence function tailored to the task and decoding scheme for successful distillation. In practice, our refined distillation approach yields 20\% speedup over standard speculative decoding across five distinct tasks, using both greedy decoding and temperature sampling. Furthermore, we extend the concept of lossless speculative decoding to incorporate a lenience factor in the rejection sampling step, offering fine-grained control over the trade-off between quality and latency in lossy decoding. Finally, adopting a strategy of "distilling for performance first and distillation for speculative decoding second" enables a remarkable 8x reduction in latency with minimal performance compromise, compared to no distillation and speculative decoding baseline.
View details
Language Model Cascades: Token-Level Uncertainty And Beyond
Neha Gupta
Aditya Menon
International Conference on Learning Representations (2024)
Preview abstract
Recent advances in language model (LM) design has yielded a series of models with remarkably improved quality on complex NLP tasks, but significantly in-creased inference cost. A simple strategy to achieve more favourable cost-quality tradeoffs is cascading: here, a small model is invoked for most “easy” instances, while a large model is invoked for a few “hard” instances. Typically, “easy” in-stances are those where the small model has high confidence in its prediction.While the principles underpinning effective cascading are well-studied for classification problems, a similar understanding is lacking for generative tasks. The ex-tension of simple ”Chow” rule which defers based on the probability of predicting an answer is not straightforward for generative tasks where the number of output tokens is variable. Moreover, LMs are known to suffer from length bias where longer answers are penalized more as compared to shorter answers which complicates things further. In this work, we initiate a systematic study of deferral rules for cascades for language models. For example, how does one best summarise model confidence across a variable number of output tokens? We show experimentally that there is no one straight forward extension of probability based uncertainty for LMs which works well across all tasks. Via experiments on a range of bench-marks with FLAN-T5 models, we find that incorporating token-level uncertainty can significantly improve the cost-quality tradeoff of cascades. We further show that incorporating embeddings from the smaller model and intermediate layer embeddings from the larger model can further boost performance
View details
Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines
Yuchen Li
Alexandre Kirchmeyer
Aashay Mehta
Yilong Qin
Andrej Risteski
International Conference on Machine Learning (2024) (to appear)
Preview abstract
Autoregressive language models are the currently dominant paradigm for text generation, however they have some fundamental limitations that cannot be remedied by scale ---for example inherently sequential and unidirectional generation. While alternate classes of models have been explored, we have limited mathematical understanding of their fundamental power and limitations. In this paper we focus on Generative Masked Language Models (GMLMs), a non-autoregressive paradigm in which we train a model to fit conditional probabilities of the data distribution via masking, which are subsequently used as inputs to a Markov Chain to draw samples from the model. These models empirically strike a promising speed-quality trade-off as each step can be typically parallelized by decoding the entire sequence in parallel. We develop a mathematical framework for analyzing and improving such models which sheds light on questions of sample complexity and inference speed and quality. Empirically, we adapt the T5 model for iteratively-refined parallel decoding, achieving 2-3x speedup in machine translation with minimal sacrifice in quality compared with autoregressive models. We run careful ablation experiments to give recommendations on key design choices, and make fine-grained observations on the common error modes in connection with our theory. Our mathematical analyses and empirical observations characterize both potentials and limitations of this approach, and can be applied to future works on improving understanding and performance of GMLMs.
View details
MarkovGen: Structured Prediction for Efficient Text-to-Image Generation
Sadeep Jayasumana
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2024)
Preview abstract
Modern text-to-image generation models produce high-quality images that are both photorealistic and faithful to the text prompts. However, this quality comes at significant computational cost: nearly all of these models are iterative and require running sampling multiple times with large models. This iterative process is needed to ensure that different regions of the image are not only aligned with the text prompt, but also compatible with each other. In this work, we propose a light-weight approach to achieving this compatibility between different regions of an image, using a Markov Random Field (MRF) model. We demonstrate the effectiveness of this method on top of the latent token-based Muse text-to-image model. The MRF richly encodes the compatibility among image tokens at different spatial locations to improve quality and significantly reduce the required number of Muse sampling steps. Inference with the MRF is significantly cheaper, and its parameters can be quickly learned through back-propagation by modeling MRF inference as a differentiable neural-network layer. Our full model, MarkovGen, uses this proposed MRF model to both speed up Muse by 1.5X and produce higher quality images by decreasing undesirable image artifacts.
View details
USTAD: Unified Single-model Training Achieving Diverse Scores for Information Retrieval
Veeru Sadhanala
Sadeep Jayasumana
Aditya Menon
Rob Fergus
International Conference on Machine Learning (ICML) (2024)
Preview abstract
Modern information retrieval (IR) systems consists of multiple stages like retrieval and ranking. Transformers are employed across these different IR stages, achieving state-of-the-art performance, but each model is trained separately leading to complex pipelines and increased cost for maintaining multiple models. The apparent need for separate models is due to different input/output semantics at different stages. In this paper, we challenge this tradition of using separate models as transformers are very expressive models and ask the question would changing just score function suffice? We present a new unified approach - USTAD - to train a single network that can provide powerful ranking scores as cross-encoder (CE) as well as factorized embeddings for large-scale retrieval as a dual-encoder (DE). Empirically, we find a single USTAD model to be competitive to separate ranking CE and retrieval DE models. Furthermore, USTAD enables new distillation techniques, significantly improving CE to DE distillations. Also using USTAD teacher, we can deploy novel asymmetric architectures for student models which realizes better embedding alignment without increasing online inference cost. On standard benchmarks like MSMARCO, we show that our approach successfully distills from both dual-encoder (DE) and cross-encoder (CE) teacher models to 1/10th size asymmetric students that can retain 95-97% of the teacher performance.
View details
Think before you speak: Training language models with pause tokens
Sachin Goyal
Ziwei Ji
Aditya Menon
Vaishnavh Nagarajan
International Conference on Learning Representations (ICLR) (2024)
Preview abstract
The present-day language model generates its response by producing a series of tokens in immediate succession: the $K+1$th token is an outcome of manipulating exactly $K$ hidden values in each layer corresponding to each of the $K$ previous tokens. Is it possible to somehow allow the model to manipulate more hidden values before committing to an answer? If yes, would this help? We explore these questions by training models with learnable \textit{pause} tokens. Besides feeding the usual prefix to the model, our idea is to feed the model with an additional sequence of pause tokens. On these tokens, the model's output is ignored all the way until the last pause token, where we begin extracting the answer. We explore this idea of ``delayed answering'' in a 1B model, where we consider both pre-training and/or fine-tuning with pause tokens. We find that while merely finetuning a standard model is not very helpful, pause-pretrained models shows promise on some downstream tasks such as GSM (reasoning) and Squad, CommonSenseQA and Lambada (question-answering tasks). We also conduct various ablations to explore the effect of the number of pause tokens. While our work takes a preliminary exploration in delayed computations for language models by focusing on a 1B model, we hope it inspires future work that can make this idea practically feasible without pre-training and for models trained with other pretraining objectives and other sizes.
View details
Rethinking FID: Towards a Better Evaluation Metric for Image Generation
Sadeep Jayasumana
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2024)
Preview abstract
As with many machine learning problems, the progress of image generation methods hinges on good evaluation metrics. One of the most popular is the Frechet Inception Distance (FID). FID estimates the distance between a distribution of Inception-v3 features of real images, and those of images generated by the algorithm. We highlight important drawbacks of FID: Inception's poor representation of the rich and varied content generated by modern text-to-image models, incorrect normality assumptions, and poor sample complexity. We call for a reevaluation of FID's use as the primary quality metric for generated images. We empirically demonstrate that FID contradicts human raters, it does not reflect gradual improvement of iterative text-to-image models, it does not capture distortion levels, and that it produces inconsistent results when varying the sample size. We also propose an alternative new metric, CMMD, based on richer CLIP embeddings and the maximum mean discrepancy distance with the Gaussian RBF kernel. It is an unbiased estimator that does not make any assumptions on the probability distribution of the embeddings and is sample efficient. Through extensive experiments and analysis, we demonstrate that FID-based evaluations of text-to-image models may be unreliable, and that CMMD offers a more robust and reliable assessment of image quality.
View details
Preview abstract
The approximate nearest neighbor (ANN) search problem is fundamental to efficiently serving many real-world machine learning applications. A number of techniques have been developed for ANN search that are efficient, accurate, and scalable. However, such techniques typically have a number of parameters that affect the speed-recall tradeoff, and exhibit poor performance when such parameters aren't properly set. Tuning these parameters has traditionally been a manual process, demanding in-depth knowledge of the underlying search algorithm. This is becoming an increasingly unrealistic demand as ANN search grows in popularity. To tackle this obstacle to ANN adoption, this work proposes a constrained optimization-based approach to tuning quantization-based ANN algorithms. Our technique takes just a desired search cost or recall as input, and then generates tunings that, empirically, are very close to the speed-recall Pareto frontier and give leading performance on standard benchmarks.
View details
Supervision complexity and its role in knowledge distillation
Hrayr Harutyunyan
Aditya Krishna Menon
International Conference on Learning Representations (ICLR) (2023)
Preview abstract
Knowledge distillation is a popular method of compressing a large teacher model (or an ensemble of models) to a more compact student model. While empirically effective, there is limited understanding of why distillation helps, and how to improve it to transfer richer knowledge from the teacher to student. In this paper, we propose a new online distillation algorithm that applies distillation using a sequence of teacher models, corresponding to different checkpoints during teacher training. Intuitively, this gradually increases the complexity of the target functions that the student model is asked to mimic. Formally, we establish generalization bounds that explicate how the target label complexity can benefit the student. We empirically demonstrate that online distillation can significantly improve over regular offline distillation, particularly in scenarios where there is a large teacher-student capacity gap.
View details
