ML libraries, often written in architecture-specific programming languages (ASPLs) that target domain-specific architectures, are key to efficient ML systems. However, writing these high-performance ML libraries is challenging because it requires expert knowledge of both ML algorithms and the ASPL. Large language models (LLMs), on the other hand, have shown general coding capabilities. However, challenges remain when using LLMs for generating ML libraries using ASPLs because 1) this task is complicated even for human experts and 2) there are limited code examples due to the esoteric and evolving nature of ASPLs. We present an adaptive self-improvement agentic system that enables LLMs to perform such complex reasoning under limited data by iteratively improving their capability through self-generated experience. In order to evaluate the effectiveness of our system, we construct a benchmark of a typical ML library and generate ASPL code with both open and closed-source LLMs on this benchmark. Our results show improvements of up to $3.9\times$ over a baseline single LLM.

Large language models (LLMs) have shown promise in transforming machine learning research, yet their capability to faithfully implement genuinely novel ideas from recent research papers—ideas unseen during pretraining—remains unclear. We introduce ResearchCodeBench, a benchmark that evaluates LLMs’ ability to translate cutting-edge ML contributions from top 2024-2025 research papers into executable code. We assessed 30+ proprietary and open-source LLMs, finding that even the best models correctly implement less than 40% of the code. We present empirical findings on performance comparison, contamination, and error patterns. By providing a rigorous evaluation platform, ResearchCodeBench enables continuous understanding and advancement of LLM-driven innovation in research code generation.

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.

The advent of large language models (LLMs) offers unprecedented opportunities to reimagine peer review beyond the constraints of traditional workflows. Despite these opportunities, prior efforts have largely focused on replicating traditional review workflows with LLMs serving as direct substitutes for human reviewers, while limited attention has been given to exploring new paradigms that fundamentally rethink how LLMs can participate in the academic review process. In this paper, we introduce and explore a novel mechanism that employs LLM agents to perform pairwise comparisons among manuscripts instead of individual scoring. By aggregating outcomes from substantial pairwise evaluations, this approach enables a more accurate and robust measure of relative manuscript quality. Our experiments demonstrate that this comparative approach significantly outperforms traditional rating-based methods in identifying high-impact papers. However, our analysis also reveals emergent biases in the selection process, notably a reduced novelty in research topics and an increased institutional imbalance. These findings highlight both the transformative potential of rethinking peer review with LLMs and critical challenges that future systems must address to ensure equity and diversity.

Advances in machine learning are closely tied to the creation of datasets. While data documentation is widely recognized as essential to the reliability, reproducibility, and transparency of ML, we lack a systematic empirical understanding of current dataset documentation practices. To shed light on this question, here we take Hugging Face - one of the largest platforms for sharing and collaborating on ML models and datasets - as a prominent case study. By analyzing all 7,433 dataset documentation on Hugging Face, our investigation provides an overview of the Hugging Face dataset ecosystem and insights into dataset documentation practices, yielding 5 main findings: (1) The dataset card completion rate shows marked heterogeneity correlated with dataset popularity: While 86.0\% of the top 100 downloaded dataset cards fill out all sections suggested by Hugging Face community, only 7.9\% of dataset cards with no downloads complete all these sections. (2) A granular examination of each section within the dataset card reveals that the practitioners seem to prioritize Dataset Description and Dataset Structure sections, accounting for 36.2\% and 33.6\% of the total card length, respectively, for the most downloaded datasets. In contrast, the Considerations for Using the Data section receives the lowest proportion of content, accounting for just 2.1\% of the text. (3) By analyzing the subsections within each section and utilizing topic modeling to identify key topics, we uncover what is discussed in each section, and underscore significant themes encompassing both technical and social impacts, as well as limitations within the Considerations for Using the Data section. (4) Our findings also highlight the need for improved accessibility and reproducibility of datasets in the Usage sections. (5) In addition, our human annotation evaluation emphasizes the pivotal role of comprehensive dataset content in shaping individuals' perceptions of a dataset card's overall quality. Overall, our study offers a unique perspective on analyzing dataset documentation through large-scale data science analysis and underlines the need for more thorough dataset documentation in machine learning research.

We present an approach for estimating the fraction of text in a large corpus which is likely to be substantially modified or produced by a large language model (LLM). Our maximum likelihood model leverages expert-written and AI-generated reference texts to accurately and efficiently examine real-world LLM-use at the corpus level. We apply this approach to a case study of scientific peer review in AI conferences that took place after the release of ChatGPT: *ICLR* 2024, *NeurIPS* 2023, *CoRL* 2023 and *EMNLP* 2023. Our results suggest that between 6.5% and 16.9% of text submitted as peer reviews to these conferences could have been substantially modified by LLMs, i.e. beyond spell-checking or minor writing updates. The circumstances in which generated text occurs offer insight into user behavior: the estimated fraction of LLM-generated text is higher in reviews which report lower confidence, were submitted close to the deadline, and from reviewers who are less likely to respond to author rebuttals. We also observe corpus-level trends in generated text which may be too subtle to detect at the individual level, and discuss the implications of such trends on peer review. We call for future interdisciplinary work to examine how LLM use is changing our information and knowledge practices.

Cluster tools are widely adopted in semiconductor manufacturing. When a process module fails, a cluster tool must undergo a close-down process, transiting to an idle state. Minimizing the makespan of a close-down process is economically significant in increasing the throughput of a wafer fab. However, the optimization of a close-down process is challenging due to wafer residency time constraints. Existing linear programming models must be constructed when specific processing time and robot activity time parameters are given. To address the issue, we propose a scheduling method based on deep reinforcement learning. For this reinforcement learning algorithm, the specific states, actions, and reward functions are designed for the close-down process to follow the Markov property, and a deep Q-network is adopted to find an optimal scheduling policy. Experimental results demonstrate that, in contrast to traditional scheduling methods, the proposed method achieves an optimal policy and also excels in generalization, enabling adaptive real-time production scheduling.

Understanding the performance of machine learning (ML) models across diverse data distributions is critically important for reliable applications. Despite recent empirical studies positing a near-perfect linear correlation between in-distribution (ID) and out-of-distribution (OOD) accuracies, we empirically demonstrate that this correlation is more nuanced under subpopulation shifts. Through rigorous experimentation and analysis across a variety of datasets, models, and training epochs, we demonstrate that OOD performance often has a nonlinear correlation with ID performance in subpopulation shifts. Our findings, which contrast previous studies that have posited a linear correlation in model performance during distribution shifts, reveal a "moon shape" correlation (parabolic uptrend curve) between the test performance on the majority subpopulation and the minority subpopulation. This non-trivial nonlinear correlation holds across model architectures, hyperparameters, training durations, and the imbalance between subpopulations. Furthermore, we found that the nonlinearity of this "moon shape" is causally influenced by the degree of spurious correlations in the training data. Our controlled experiments show that stronger spurious correlation in the training data creates more nonlinear performance correlation. We provide complementary experimental and theoretical analyses for this phenomenon, and discuss its implications for ML reliability and fairness. Our work highlights the importance of understanding the nonlinear effects of model improvement on performance in different subpopulations, and has the potential to inform the development of more equitable and responsible machine learning models.

Assessing the quality and impact of individual data points is critical for improving model performance and mitigating undesirable biases within the training dataset. Several data valuation algorithms have been proposed to quantify data quality, however, there lacks a systemic and standardized benchmarking system for data valuation. In this paper, we introduce OpenDataVal, an easy-to-use and unified benchmark framework that empowers researchers and practitioners to apply and compare various data valuation algorithms. OpenDataVal provides an integrated environment that includes (i) a diverse collection of image, natural language, and tabular datasets, (ii) implementations of eleven different state-of-the-art data valuation algorithms, and (iii) a prediction model API that can import any models in scikit-learn. Furthermore, we propose four downstream machine learning tasks for evaluating the quality of data values. We perform benchmarking analysis using OpenDataVal, quantifying and comparing the efficacy of state-of-the-art data valuation approaches. We find that no single algorithm performs uniformly best across all tasks, and an appropriate algorithm should be employed for a user's downstream task. OpenDataVal is publicly available at https://opendataval.github.io with comprehensive documentation. Furthermore, we provide a leaderboard where researchers can evaluate the effectiveness of their own data valuation algorithms.

Machine learning algorithms typically assume that training and test examples are drawn from the same distribution. However, distribution shift is a common problem in real-world applications and can cause models to perform dramatically worse at test time. In this paper, we specifically consider the problems of subpopulation shifts (e.g., imbalanced data) and domain shifts. While prior works often seek to explicitly regularize internal representations or predictors of the model to be domain invariant, we instead aim to learn invariant predictors without restricting the model’s internal representations or predictors. This leads to a simple mixup-based technique which learns invariant predictors via selective augmentation called LISA. LISA selectively interpolates samples either with the same labels but different domains or with the same domain but different labels. Empirically, we study the effectiveness of LISA on nine benchmarks ranging from subpopulation shifts to domain shifts, and we find that LISA consistently outperforms other state-of-the-art methods and leads to more invariant predictors. We further analyze a linear setting and theoretically show how LISA leads to a smaller worst-group error.

We present modality gap, an intriguing geometric phenomenon of the representation space of multi-modal models. Specifically, we show that different data modalities (e.g. images and text) are embedded at arm's length in their shared representation in multi-modal models such as CLIP. Our systematic analysis demonstrates that this gap is caused by a combination of model initialization and contrastive learning optimization. In model initialization, we show empirically and theoretically that the representation of a common deep neural network is restricted to a narrow cone. As a consequence, in a multi-modal model with two encoders, the representations of the two modalities are clearly apart when the model is initialized. During optimization, contrastive learning keeps the different modalities separate by a certain distance, which is influenced by the temperature parameter in the loss function. Our experiments further demonstrate that varying the modality gap distance has a significant impact in improving the model's downstream zero-shot classification performance and fairness.

Weixin Liang, Kai-Hui Liang, Zhou Yu. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Recent advances in deep learning have made the use of large, deep neural networks with tens of millions of parameters. The sheer size of these networks imposes a challenging computational burden during inference. Existing work focuses primarily on accelerating each forward pass of a neural network. Inspired by the group testing strategy for efficient disease testing, we propose neural group testing, which accelerates by testing a group of samples in one forward pass. Groups of samples that test negative are ruled out. If a group tests positive, samples in that group are then retested adaptively. A key challenge of neural group testing is to modify a deep neural network so that it could test multiple samples in one forward pass. We propose three designs to achieve this without introducing any new parameters and evaluate their performances. We applied neural group testing in an image moderation task to detect rare but inappropriate images. We found that neural group testing can group up to 16 images in one forward pass and reduce the overall computation cost by over 73% while improving detection performance. Our code is available at https://github.com/Weixin-Liang/NeuralGroupTesting/

A major bottleneck in training end-to-end task-oriented dialog system is the lack of data. To utilize limited training data more efficiently, we propose Modular Supervision Network (MOSS), an encoder-decoder training framework that could incorporate supervision from various intermediate dialog system modules including natural language understanding, dialog state tracking, dialog policy learning and natural language generation. With only 60% of the training data, MOSS-all (i.e., MOSS with supervision from all four dialog modules) outperforms state-of-the-art models on CamRest676. Moreover, introducing modular supervision has even bigger benefits when the dialog task has a more complex dialog state and action space. With only 40% of the training data, MOSS-all outperforms the state-of-the-art model on a complex laptop network trouble shooting dataset, LaptopNetwork, that we introduced. LaptopNetwork consists of conversations between real customers and customer service agents in Chinese. Moreover, MOSS framework can accommodate dialogs that have supervision from different dialog modules at both framework level and model level. Therefore, MOSS is extremely flexible to update in real-world deployment.

Open Domain dialog system evaluation is one of the most important challenges in dialog research.Existing automatic evaluation metrics, such as BLEU are mostly referencebased.They calculate the difference between the generated response and a limited number of available references.Likert-score based self-reported user rating is widely adopted by social conversational systems, such as Amazon Alexa Prize chatbots.However, selfreported user rating suffers from bias and variance among different users.To alleviate this problem, we formulate dialog evaluation as a comparison task.We also propose an automatic evaluation model CMADE (Comparison Model for Automatic Dialog Evaluation) that automatically cleans self-reported user ratings as it trains on them.Specifically, we first use a self-supervised method to learn better dialog feature representation, and then use KNN and Shapley to remove confusing samples.Our experiments show that CMADE achieves 89.2% accuracy in the dialog comparison task.Our implementation is available at https://github.com/Weixin-Liang/ dialog_evaluation_CMADE.

Training a supervised neural network classifier typically requires many annotated training samples.Collecting and annotating a large number of data points are costly and sometimes even infeasible.Traditional annotation process uses a low-bandwidth human-machine communication interface: classification labels, each of which only provides a few bits of information.We propose Active Learning with Contrastive Explanations (ALICE), an expert-in-the-loop training framework that utilizes contrastive natural language explanations to improve data efficiency in learning.AL-ICE learns to first use active learning to select the most informative pairs of label classes to elicit contrastive natural language explanations from experts.Then it extracts knowledge from these explanations using a semantic parser.Finally, it incorporates the extracted knowledge through dynamically changing the learning model's structure.We applied ALICE in two visual recognition tasks, bird species classification and social relationship classification.We found by incorporating contrastive explanations, our models outperform baseline models that are trained with 40-100% more training data.We found that adding 1 explanation leads to similar performance gain as adding 13-30 labeled training data points.

Recent advancements in deep learning techniques facilitate intelligent-query support in diverse applications, such as content-based image retrieval and audio texturing. Unlike conventional key-based queries, these intelligent queries lack efficient indexing and require complex compute operations for feature matching. To achieve high-performance intelligent querying against massive datasets, modern computing systems employ GPUs in-conjunction with solid-state drives (SSDs) for fast data access and parallel data processing. However, our characterization with various intelligent-query workloads developed with deep neural networks (DNNs), shows that the storage I/O bandwidth is still the major bottleneck that contributes 56%--90% of the query execution time.

Access patterns over untrusted memory have long been exploited to infer sensitive information like program types or even secret keys. Most existing obfuscation solutions hide real memory accesses among a sufficiently large number of dummy memory accesses. Such solutions lead to a heavy communication overhead and more often apply to the client/server scenario instead of the CPU/memory architecture. Sporadic obfuscation solutions strive for an affordable memory bandwidth cost at the expense of security degradation. For example, they may have to obfuscate accesses over a limited range of memory space to control the overhead.