Ge Zhang

Ruibin Yuan, Yinghao Ma, Yizhi Li et al. · deepmind, mila

In the era of extensive intersection between art and Artificial Intelligence (AI), such as image generation and fiction co-creation, AI for music remains relatively nascent, particularly in music understanding. This is evident in the limited work on deep music representations, the scarcity of large-scale datasets, and the absence of a universal and community-driven benchmark. To address this issue, we introduce the Music Audio Representation Benchmark for universaL Evaluation, termed MARBLE. It aims to provide a benchmark for various Music Information Retrieval (MIR) tasks by defining a comprehensive taxonomy with four hierarchy levels, including acoustic, performance, score, and high-level description. We then establish a unified protocol based on 14 tasks on 8 public-available datasets, providing a fair and standard assessment of representations of all open-sourced pre-trained models developed on music recordings as baselines. Besides, MARBLE offers an easy-to-use, extendable, and reproducible suite for the community, with a clear statement on copyright issues on datasets. Results suggest recently proposed large-scale pre-trained musical language models perform the best in most tasks, with room for further improvement. The leaderboard and toolkit repository are published at https://marble-bm.shef.ac.uk to promote future music AI research.

Yizhi Li, Ge Zhang, Bohao Yang et al. · mila

Fairness has become a trending topic in natural language processing (NLP), which addresses biases targeting certain social groups such as genders and religions. However, regional bias in language models (LMs), a long-standing global discrimination problem, still remains unexplored. This paper bridges the gap by analysing the regional bias learned by the pre-trained language models that are broadly used in NLP tasks. In addition to verifying the existence of regional bias in LMs, we find that the biases on regional groups can be strongly influenced by the geographical clustering of the groups. We accordingly propose a HiErarchical Regional Bias evaluation method (HERB) utilising the information from the sub-region clusters to quantify the bias in pre-trained LMs. Experiments show that our hierarchical metric can effectively evaluate the regional bias with respect to comprehensive topics and measure the potential regional bias that can be propagated to downstream tasks. Our codes are available at https://github.com/Bernard-Yang/HERB.

Adam Nik, Ge Zhang, Xingran Chen et al. · mila

This paper details our participation in the Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE) workshop @ EMNLP 2022, where we take part in Subtask 1 of Shared Task 3. We approach the given task of event causality detection by proposing a self-training pipeline that follows a teacher-student classifier method. More specifically, we initially train a teacher model on the true, original task data, and use that teacher model to self-label data to be used in the training of a separate student model for the final task prediction. We test how restricting the number of positive or negative self-labeled examples in the self-training process affects classification performance. Our final results show that using self-training produces a comprehensive performance improvement across all models and self-labeled training sets tested within the task of event causality sequence classification. On top of that, we find that self-training performance did not diminish even when restricting either positive/negative examples used in training. Our code is be publicly available at https://github.com/Gzhang-umich/1CademyTeamOfCASE.

Xianjie Wu, Jian Yang, Linzheng Chai et al.

Recent advancements in Large Language Models (LLMs) have markedly enhanced the interpretation and processing of tabular data, introducing previously unimaginable capabilities. Despite these achievements, LLMs still encounter significant challenges when applied in industrial scenarios, particularly due to the increased complexity of reasoning required with real-world tabular data, underscoring a notable disparity between academic benchmarks and practical applications. To address this discrepancy, we conduct a detailed investigation into the application of tabular data in industrial scenarios and propose a comprehensive and complex benchmark TableBench, including 18 fields within four major categories of table question answering (TableQA) capabilities. Furthermore, we introduce TableLLM, trained on our meticulously constructed training set TableInstruct, achieving comparable performance with GPT-3.5. Massive experiments conducted on TableBench indicate that both open-source and proprietary LLMs still have significant room for improvement to meet real-world demands, where the most advanced model, GPT-4, achieves only a modest score compared to humans.

Yinghao Ma, Ruibin Yuan, Yizhi Li et al. · deepmind, mila

Self-supervised learning (SSL) has shown promising results in various speech and natural language processing applications. However, its efficacy in music information retrieval (MIR) still remains largely unexplored. While previous SSL models pre-trained on music recordings may have been mostly closed-sourced, recent speech models such as wav2vec2.0 have shown promise in music modelling. Nevertheless, research exploring the effectiveness of applying speech SSL models to music recordings has been limited. We explore the music adaption of SSL with two distinctive speech-related models, data2vec1.0 and Hubert, and refer to them as music2vec and musicHuBERT, respectively. We train $12$ SSL models with 95M parameters under various pre-training configurations and systematically evaluate the MIR task performances with 13 different MIR tasks. Our findings suggest that training with music data can generally improve performance on MIR tasks, even when models are trained using paradigms designed for speech. However, we identify the limitations of such existing speech-oriented designs, especially in modelling polyphonic information. Based on the experimental results, empirical suggestions are also given for designing future musical SSL strategies and paradigms.

Zihao Deng, Yinghao Ma, Yudong Liu et al.

Large Language Models (LLMs) have shown immense potential in multimodal applications, yet the convergence of textual and musical domains remains not well-explored. To address this gap, we present MusiLingo, a novel system for music caption generation and music-related query responses. MusiLingo employs a single projection layer to align music representations from the pre-trained frozen music audio model MERT with a frozen LLM, bridging the gap between music audio and textual contexts. We train it on an extensive music caption dataset and fine-tune it with instructional data. Due to the scarcity of high-quality music Q&A datasets, we created the MusicInstruct (MI) dataset from captions in the MusicCaps datasets, tailored for open-ended music inquiries. Empirical evaluations demonstrate its competitive performance in generating music captions and composing music-related Q&A pairs. Our introduced dataset enables notable advancements beyond previous ones.

Tianyu Zheng, Ge Zhang, Tianhao Shen et al.

The introduction of large language models has significantly advanced code generation. However, open-source models often lack the execution capabilities and iterative refinement of advanced systems like the GPT-4 Code Interpreter. To address this, we introduce OpenCodeInterpreter, a family of open-source code systems designed for generating, executing, and iteratively refining code. Supported by Code-Feedback, a dataset featuring 68K multi-turn interactions, OpenCodeInterpreter integrates execution and human feedback for dynamic code refinement. Our comprehensive evaluation of OpenCodeInterpreter across key benchmarks such as HumanEval, MBPP, and their enhanced versions from EvalPlus reveals its exceptional performance. Notably, OpenCodeInterpreter-33B achieves an accuracy of 83.2 (76.4) on the average (and plus versions) of HumanEval and MBPP, closely rivaling GPT-4's 84.2 (76.2) and further elevates to 91.6 (84.6) with synthesized human feedback from GPT-4. OpenCodeInterpreter brings the gap between open-source code generation models and proprietary systems like GPT-4 Code Interpreter.

Jiawei Guo, Ziming Li, Xueling Liu et al.

Large Language Models (LLMs) for code are rapidly evolving, with code editing emerging as a critical capability. We introduce CodeEditorBench, an evaluation framework designed to rigorously assess the performance of LLMs in code editing tasks, including debugging, translating, polishing, and requirement switching. Unlike existing benchmarks focusing solely on code generation, CodeEditorBench emphasizes real-world scenarios and practical aspects of software development. We curate diverse coding challenges and scenarios from five sources, covering various programming languages, complexity levels, and editing tasks. Evaluation of 19 LLMs reveals that closed-source models (particularly Gemini-Ultra and GPT-4), outperform open-source models in CodeEditorBench, highlighting differences in model performance based on problem types and prompt sensitivities. CodeEditorBench aims to catalyze advancements in LLMs by providing a robust platform for assessing code editing capabilities. We will release all prompts and datasets to enable the community to expand the dataset and benchmark emerging LLMs. By introducing CodeEditorBench, we contribute to the advancement of LLMs in code editing and provide a valuable resource for researchers and practitioners.

Alexander Zhang, Marcus Dong, Jiaheng Liu et al.

The critique capacity of Large Language Models (LLMs) is essential for reasoning abilities, which can provide necessary suggestions (e.g., detailed analysis and constructive feedback). Therefore, how to evaluate the critique capacity of LLMs has drawn great attention and several critique benchmarks have been proposed. However, existing critique benchmarks usually have the following limitations: (1). Focusing on diverse reasoning tasks in general domains and insufficient evaluation on code tasks (e.g., only covering code generation task), where the difficulty of queries is relatively easy (e.g., the code queries of CriticBench are from Humaneval and MBPP). (2). Lacking comprehensive evaluation from different dimensions. To address these limitations, we introduce a holistic code critique benchmark for LLMs called CodeCriticBench. Specifically, our CodeCriticBench includes two mainstream code tasks (i.e., code generation and code QA) with different difficulties. Besides, the evaluation protocols include basic critique evaluation and advanced critique evaluation for different characteristics, where fine-grained evaluation checklists are well-designed for advanced settings. Finally, we conduct extensive experimental results of existing LLMs, which show the effectiveness of CodeCriticBench.

David Ma, Huaqing Yuan, Xingjian Wang et al.

Although long-video understanding demands that models capture hierarchical temporal information -- from clip (seconds) and shot (tens of seconds) to event (minutes) and story (hours) -- existing benchmarks either neglect this multi-scale design or scatter scale-specific questions across different videos, preventing direct comparison of model performance across timescales on the same content. To address this, we introduce ScaleLong, the first benchmark to disentangle these factors by embedding questions targeting four hierarchical timescales -- clip (seconds), shot (tens of seconds), event (minutes), and story (hours) -- all within the same video content. This within-content multi-timescale questioning design enables direct comparison of model performance across timescales on identical videos. ScaleLong features 269 long videos (avg.\ 86\,min) from 5 main categories and 36 sub-categories, with 4--8 carefully designed questions, including at least one question for each timescale. Evaluating 23 MLLMs reveals a U-shaped performance curve, with higher accuracy at the shortest and longest timescales and a dip at intermediate levels. Furthermore, ablation studies show that increased visual token capacity consistently enhances reasoning across all timescales. ScaleLong offers a fine-grained, multi-timescale benchmark for advancing MLLM capabilities in long-video understanding. The code and dataset are available https://github.com/multimodal-art-projection/ScaleLong.

Siming Huang, Tianhao Cheng, J. K. Liu et al.

Large language models (LLMs) for code have become indispensable in various domains, including code generation, reasoning tasks and agent systems. While open-access code LLMs are increasingly approaching the performance levels of proprietary models, high-quality code LLMs suitable for rigorous scientific investigation, particularly those with reproducible data processing pipelines and transparent training protocols, remain limited. The scarcity is due to various challenges, including resource constraints, ethical considerations, and the competitive advantages of keeping models advanced. To address the gap, we introduce OpenCoder, a top-tier code LLM that not only achieves performance comparable to leading models but also serves as an "open cookbook" for the research community. Unlike most prior efforts, we release not only model weights and inference code, but also the reproducible training data, complete data processing pipeline, rigorous experimental ablation results, and detailed training protocols for open scientific research. Through this comprehensive release, we identify the key ingredients for building a top-tier code LLM: (1) code optimized heuristic rules for data cleaning and methods for data deduplication, (2) recall of text corpus related to code and (3) high-quality synthetic data in both annealing and supervised fine-tuning stages. By offering this level of openness, we aim to broaden access to all aspects of a top-tier code LLM, with OpenCoder serving as both a powerful model and an open foundation to accelerate research, and enable reproducible advancements in code AI.

Ashley Zhang

Standard approaches to causal inference, such as Outcome Regression and Inverse Probability Weighted Regression Adjustment (IPWRA), are typically derived through the lens of missing data imputation and identification theory. In this work, we unify these methods from a Machine Learning perspective, reframing ATE estimation as a \textit{domain adaptation problem under distribution shift}. We demonstrate that the canonical Hajek estimator is a special case of IPWRA restricted to a constant hypothesis class, and that IPWRA itself is fundamentally Importance-Weighted Empirical Risk Minimization designed to correct for the covariate shift between the treated sub-population and the target population. Leveraging this unified framework, we critically examine the optimization objectives of Doubly Robust estimators. We argue that standard methods enforce \textit{sufficient but not necessary} conditions for consistency by requiring outcome models to be individually unbiased. We define the true "ATE Risk Function" and show that minimizing it requires only that the biases of the treated and control models structurally cancel out. Exploiting this insight, we propose the \textbf{Joint Robust Estimator (JRE)}. Instead of treating propensity estimation and outcome modeling as independent stages, JRE utilizes bootstrap-based uncertainty quantification of the propensity score to train outcome models jointly. By optimizing for the expected ATE risk over the distribution of propensity scores, JRE leverages model degrees of freedom to achieve robustness against propensity misspecification. Simulation studies demonstrate that JRE achieves up to a 15\% reduction in MSE compared to standard IPWRA in finite-sample regimes with misspecified outcome models.

Jiaheng Liu, Ken Deng, Congnan Liu et al.

Repository-level code completion has drawn great attention in software engineering, and several benchmark datasets have been introduced. However, existing repository-level code completion benchmarks usually focus on a limited number of languages (<5), which cannot evaluate the general code intelligence abilities across different languages for existing code Large Language Models (LLMs). Besides, the existing benchmarks usually report overall average scores of different languages, where the fine-grained abilities in different completion scenarios are ignored. Therefore, to facilitate the research of code LLMs in multilingual scenarios, we propose a massively multilingual repository-level code completion benchmark covering 18 programming languages (called M2RC-EVAL), and two types of fine-grained annotations (i.e., bucket-level and semantic-level) on different completion scenarios are provided, where we obtain these annotations based on the parsed abstract syntax tree. Moreover, we also curate a massively multilingual instruction corpora M2RC- INSTRUCT dataset to improve the repository-level code completion abilities of existing code LLMs. Comprehensive experimental results demonstrate the effectiveness of our M2RC-EVAL and M2RC-INSTRUCT.

Shangda Wu, Yashan Wang, Ruibin Yuan et al.

Challenges in managing linguistic diversity and integrating various musical modalities are faced by current music information retrieval systems. These limitations reduce their effectiveness in a global, multimodal music environment. To address these issues, we introduce CLaMP 2, a system compatible with 101 languages that supports both ABC notation (a text-based musical notation format) and MIDI (Musical Instrument Digital Interface) for music information retrieval. CLaMP 2, pre-trained on 1.5 million ABC-MIDI-text triplets, includes a multilingual text encoder and a multimodal music encoder aligned via contrastive learning. By leveraging large language models, we obtain refined and consistent multilingual descriptions at scale, significantly reducing textual noise and balancing language distribution. Our experiments show that CLaMP 2 achieves state-of-the-art results in both multilingual semantic search and music classification across modalities, thus establishing a new standard for inclusive and global music information retrieval.

Shilong Li, Xingyuan Bu, Wenjie Wang et al.

AI agents with advanced reasoning and tool use capabilities have demonstrated impressive performance in web browsing for deep search. While existing benchmarks such as BrowseComp evaluate these browsing abilities, they primarily focus on textual information, overlooking the prevalence of multimodal content. To bridge this gap, we introduce MM-BrowseComp, a novel benchmark comprising 224 challenging, hand-crafted questions specifically designed to assess agents' multimodal retrieval and reasoning capabilities. These questions often incorporate images in prompts, and crucial information encountered during the search and reasoning process may also be embedded within images or videos on webpages. Consequently, methods relying solely on text prove insufficient for our benchmark. Additionally, we provide a verified checklist for each question, enabling fine-grained analysis of multimodal dependencies and reasoning paths. Our comprehensive evaluation of state-of-the-art models on MM-BrowseComp reveals that even top models like OpenAI o3 with tools achieve only 29.02\% accuracy, highlighting the suboptimal multimodal capabilities and lack of native multimodal reasoning in current models.

Jian Yang, Xianglong Liu, Weifeng Lv et al.

Large language models (LLMs) have fundamentally transformed automated software development by enabling direct translation of natural language descriptions into functional code, driving commercial adoption through tools like Github Copilot (Microsoft), Cursor (Anysphere), Trae (ByteDance), and Claude Code (Anthropic). While the field has evolved dramatically from rule-based systems to Transformer-based architectures, achieving performance improvements from single-digit to over 95\% success rates on benchmarks like HumanEval. In this work, we provide a comprehensive synthesis and practical guide (a series of analytic and probing experiments) about code LLMs, systematically examining the complete model life cycle from data curation to post-training through advanced prompting paradigms, code pre-training, supervised fine-tuning, reinforcement learning, and autonomous coding agents. We analyze the code capability of the general LLMs (GPT-4, Claude, LLaMA) and code-specialized LLMs (StarCoder, Code LLaMA, DeepSeek-Coder, and QwenCoder), critically examining the techniques, design decisions, and trade-offs. Further, we articulate the research-practice gap between academic research (e.g., benchmarks and tasks) and real-world deployment (e.g., software-related code tasks), including code correctness, security, contextual awareness of large codebases, and integration with development workflows, and map promising research directions to practical needs. Last, we conduct a series of experiments to provide a comprehensive analysis of code pre-training, supervised fine-tuning, and reinforcement learning, covering scaling law, framework selection, hyperparameter sensitivity, model architectures, and dataset comparisons.

Yizhi Li, Ruibin Yuan, Ge Zhang et al.

Self-supervised learning (SSL) has recently emerged as a promising paradigm for training generalisable models on large-scale data in the fields of vision, text, and speech. Although SSL has been proven effective in speech and audio, its application to music audio has yet to be thoroughly explored. This is partially due to the distinctive challenges associated with modelling musical knowledge, particularly tonal and pitched characteristics of music. To address this research gap, we propose an acoustic Music undERstanding model with large-scale self-supervised Training (MERT), which incorporates teacher models to provide pseudo labels in the masked language modelling (MLM) style acoustic pre-training. In our exploration, we identified an effective combination of teacher models, which outperforms conventional speech and audio approaches in terms of performance. This combination includes an acoustic teacher based on Residual Vector Quantisation - Variational AutoEncoder (RVQ-VAE) and a musical teacher based on the Constant-Q Transform (CQT). Furthermore, we explore a wide range of settings to overcome the instability in acoustic language model pre-training, which allows our designed paradigm to scale from 95M to 330M parameters. Experimental results indicate that our model can generalise and perform well on 14 music understanding tasks and attain state-of-the-art (SOTA) overall scores.

Ge Zhang, Mike A. Merrill, Yang Liu et al.

Large scale analysis of source code, and in particular scientific source code, holds the promise of better understanding the data science process, identifying analytical best practices, and providing insights to the builders of scientific toolkits. However, large corpora have remained unanalyzed in depth, as descriptive labels are absent and require expert domain knowledge to generate. We propose a novel weakly supervised transformer-based architecture for computing joint representations of code from both abstract syntax trees and surrounding natural language comments. We then evaluate the model on a new classification task for labeling computational notebook cells as stages in the data analysis process from data import to wrangling, exploration, modeling, and evaluation. We show that our model, leveraging only easily-available weak supervision, achieves a 38% increase in accuracy over expert-supplied heuristics and outperforms a suite of baselines. Our model enables us to examine a set of 118,000 Jupyter Notebooks to uncover common data analysis patterns. Focusing on notebooks with relationships to academic articles, we conduct the largest ever study of scientific code and find that notebook composition correlates with the citation count of corresponding papers.