Debojyoti Dutta

Luis Oala, Manil Maskey, Lilith Bat-Leah et al. · mit

Drawing from discussions at the inaugural DMLR workshop at ICML 2023 and meetings prior, in this report we outline the relevance of community engagement and infrastructure development for the creation of next-generation public datasets that will advance machine learning science. We chart a path forward as a collective effort to sustain the creation and maintenance of these datasets and methods towards positive scientific, societal and business impact.

Aditya Bharat Soni, Rajat Ghosh, Vaishnavi Bhargava et al.

Software testing is crucial for ensuring the correctness and reliability of software systems. Automated generation of issue reproduction tests from natural language issue descriptions enhances developer productivity by simplifying root cause analysis, promotes test-driven development -- "test first, write code later", and can be used for improving the effectiveness of automated issue resolution systems like coding agents. Existing methods proposed for this task predominantly rely on closed-source LLMs, with limited exploration of open models. To address this, we propose SWE-Tester -- a novel pipeline for training open-source LLMs to generate issue reproduction tests. First, we curate a high-quality training dataset of 41K instances from 2.6K open-source GitHub repositories and use it to train LLMs of varying sizes and families. The fine-tuned models achieve absolute improvements of up to 10\% in success rate and 21\% in change coverage on SWT-Bench Verified. Further analysis shows consistent improvements with increased inference-time compute, more data, and larger models. These results highlight the effectiveness of our framework for advancing open-source LLMs in this domain.

Yashshi Pipalani, Hritik Raj, Rajat Ghosh et al.

Training data imbalance poses a major challenge for code LLMs. Most available data heavily over represents raw opensource code while underrepresenting broader software engineering tasks, especially in low resource languages like Golang. As a result, models excel at code autocompletion but struggle with real world developer workflows such as unit test generation. To address this gap, we introduce GO UT Bench, a benchmark dataset of 5264 pairs of code and unit tests, drawn from 10 permissively licensed Golang repositories spanning diverse domain. We evaluate its effectiveness as a fine tuning dataset across two LLM families i.e. mixture of experts and dense decoders. Our results show that finetuned models outperform their base counterparts on more than 75% of benchmark tasks.

Rajat Ghosh, Debojyoti Dutta

Training action space selection for reinforcement learning (RL) is conflict-prone due to complex state-action relationships. To address this challenge, this paper proposes a Shapley-inspired methodology for training action space categorization and ranking. To reduce exponential-time shapley computations, the methodology includes a Monte Carlo simulation to avoid unnecessary explorations. The effectiveness of the methodology is illustrated using a cloud infrastructure resource tuning case study. It reduces the search space by 80\% and categorizes the training action sets into dispensable and indispensable groups. Additionally, it ranks different training actions to facilitate high-performance yet cost-efficient RL model design. The proposed data-driven methodology is extensible to different domains, use cases, and reinforcement learning algorithms.

Vaishnavi Bhargava, Rajat Ghosh, Debojyoti Dutta

We introduce CPP-UT-Bench, a benchmark dataset to measure C++ unit test generation capability of a large language model (LLM). CPP-UT-Bench aims to reflect a broad and diverse set of C++ codebases found in the real world. The dataset includes 2,653 {code, unit test} pairs drawn from 14 different opensource C++ codebases spanned across nine diverse domains including machine learning, software testing, parsing, standard input-output, data engineering, logging, complete expression evaluation, key value storage, and server protocols. We demonstrated the effectiveness of CPP-UT-Bench as a benchmark dataset through extensive experiments in in-context learning, parameter-efficient fine-tuning (PEFT), and full-parameter fine-tuning. We also discussed the challenges of the dataset compilation and insights we learned from in-context learning and fine-tuning experiments. Besides the CPP-UT-Bench dataset and data compilation code, we are also offering the fine-tuned model weights for further research. For nine out of ten experiments, our fine-tuned LLMs outperformed the corresponding base models by an average of more than 70%.

Kristen Pereira, Neelabh Sinha, Rajat Ghosh et al.

Recent advances in frontier large language models have enabled code review agents that operate in open-ended, reasoning-intensive settings. However, the lack of standardized benchmarks and granular evaluation protocols makes it difficult to assess behavior of code review agents beyond coarse success metrics, particularly for tasks where false positives are costly. To address this gap, we introduce CR-Bench, a benchmarking dataset, and CR-Evaluator, a fine-grained evaluation pipeline for code review agents. Using these tools, we conduct a preliminary study evaluating both a single-shot agent and a Reflexion-based agent across two frontier models. We find that code review agents can exhibit a low signal-to-noise ratio when designed to identify all hidden issues, obscuring true progress and developer productivity when measured solely by resolution rates. Our analysis identifies the hidden trade-off between issue resolution and spurious findings, revealing a frontier that constrains effective agent design. Together, CR-Bench and CR-Evaluator provide a timely foundation for studying and developing code review agents as LLM-based systems transition from controlled benchmarks to real-world software engineering workflows.

Rajat Ghosh, Debojyoti Dutta

Numerous offline and model-based reinforcement learning systems incorporate world models to emulate the inherent environments. A world model is particularly important in scenarios where direct interactions with the real environment is costly, dangerous, or impractical. The efficacy and interpretability of such world models are notably contingent upon the quality of the underlying training data. In this context, we introduce Action Shapley as an agnostic metric for the judicious and unbiased selection of training data. To facilitate the computation of Action Shapley, we present a randomized dynamic algorithm specifically designed to mitigate the exponential complexity inherent in traditional Shapley value computations. Through empirical validation across five data-constrained real-world case studies, the algorithm demonstrates a computational efficiency improvement exceeding 80\% in comparison to conventional exponential time computations. Furthermore, our Action Shapley-based training data selection policy consistently outperforms ad-hoc training data selection.

Yongjun He, Roger Waleffe, Zhichao Han et al.

Many modern machine learning (ML) methods rely on embedding models to learn vector representations (embeddings) for a set of entities (embedding tables). As increasingly diverse ML applications utilize embedding models and embedding tables continue to grow in size and number, there has been a surge in the ad-hoc development of specialized frameworks targeted to train large embedding models for specific tasks. Although the scalability issues that arise in different embedding model training tasks are similar, each of these frameworks independently reinvents and customizes storage components for specific tasks, leading to substantial duplicated engineering efforts in both development and deployment. This paper presents MLKV, an efficient, extensible, and reusable data storage framework designed to address the scalability challenges in embedding model training, specifically data stall and staleness. MLKV augments disk-based key-value storage by democratizing optimizations that were previously exclusive to individual specialized frameworks and provides easy-to-use interfaces for embedding model training tasks. Extensive experiments on open-source workloads, as well as applications in eBay's payment transaction risk detection and seller payment risk detection, show that MLKV outperforms offloading strategies built on top of industrial-strength key-value stores by 1.6-12.6x. MLKV is open-source at https://github.com/llm-db/MLKV.

Bertie Vidgen, Adarsh Agrawal, Ahmed M. Ahmed et al. · deepmind, oxford

This paper introduces v0.5 of the AI Safety Benchmark, which has been created by the MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to assess the safety risks of AI systems that use chat-tuned language models. We introduce a principled approach to specifying and constructing the benchmark, which for v0.5 covers only a single use case (an adult chatting to a general-purpose assistant in English), and a limited set of personas (i.e., typical users, malicious users, and vulnerable users). We created a new taxonomy of 13 hazard categories, of which 7 have tests in the v0.5 benchmark. We plan to release version 1.0 of the AI Safety Benchmark by the end of 2024. The v1.0 benchmark will provide meaningful insights into the safety of AI systems. However, the v0.5 benchmark should not be used to assess the safety of AI systems. We have sought to fully document the limitations, flaws, and challenges of v0.5. This release of v0.5 of the AI Safety Benchmark includes (1) a principled approach to specifying and constructing the benchmark, which comprises use cases, types of systems under test (SUTs), language and context, personas, tests, and test items; (2) a taxonomy of 13 hazard categories with definitions and subcategories; (3) tests for seven of the hazard categories, each comprising a unique set of test items, i.e., prompts. There are 43,090 test items in total, which we created with templates; (4) a grading system for AI systems against the benchmark; (5) an openly available platform, and downloadable tool, called ModelBench that can be used to evaluate the safety of AI systems on the benchmark; (6) an example evaluation report which benchmarks the performance of over a dozen openly available chat-tuned language models; (7) a test specification for the benchmark.

Shaona Ghosh, Heather Frase, Adina Williams et al. · deepmind, stanford

The rapid advancement and deployment of AI systems have created an urgent need for standard safety-evaluation frameworks. This paper introduces AILuminate v1.0, the first comprehensive industry-standard benchmark for assessing AI-product risk and reliability. Its development employed an open process that included participants from multiple fields. The benchmark evaluates an AI system's resistance to prompts designed to elicit dangerous, illegal, or undesirable behavior in 12 hazard categories, including violent crimes, nonviolent crimes, sex-related crimes, child sexual exploitation, indiscriminate weapons, suicide and self-harm, intellectual property, privacy, defamation, hate, sexual content, and specialized advice (election, financial, health, legal). Our method incorporates a complete assessment standard, extensive prompt datasets, a novel evaluation framework, a grading and reporting system, and the technical as well as organizational infrastructure for long-term support and evolution. In particular, the benchmark employs an understandable five-tier grading scale (Poor to Excellent) and incorporates an innovative entropy-based system-response evaluation. In addition to unveiling the benchmark, this report also identifies limitations of our method and of building safety benchmarks generally, including evaluator uncertainty and the constraints of single-turn interactions. This work represents a crucial step toward establishing global standards for AI risk and reliability evaluation while acknowledging the need for continued development in areas such as multiturn interactions, multimodal understanding, coverage of additional languages, and emerging hazard categories. Our findings provide valuable insights for model developers, system integrators, and policymakers working to promote safer AI deployment.

Datta Nimmaturi, Vaishnavi Bhargava, Rajat Ghosh et al.

Fine-tuning large language models (LLMs) for reasoning tasks using reinforcement learning methods like Group Relative Policy Optimization (GRPO) is computationally expensive. To address this, we propose a predictive framework that models training dynamics and helps optimize resource usage. Through experiments on Llama and Qwen models (3B 8B), we derive an empirical scaling law based on model size, initial performance, and training progress. This law predicts reward trajectories and identifies three consistent training phases: slow start, rapid improvement, and plateau. We find that training beyond certain number of an epoch offers little gain, suggesting earlier stopping can significantly reduce compute without sacrificing performance. Our approach generalizes across model types, providing a practical guide for efficient GRPO-based fine-tuning.

Amrit Khera, Rajat Ghosh, Debojyoti Dutta

LLM alignment ensures that large language models behave safely and effectively by aligning their outputs with human values, goals, and intentions. Aligning LLMs employ huge amounts of data, computation, and time. Moreover, curating data with human feedback is expensive and takes time. Recent research depicts the benefit of data engineering in the fine-tuning and pre-training paradigms to bring down such costs. However, alignment differs from the afore-mentioned paradigms and it is unclear if data efficient alignment is feasible. In this work, we first aim to understand how the performance of LLM alignment scales with data. We find out that LLM alignment performance follows an exponential plateau pattern which tapers off post a rapid initial increase. Based on this, we identify data subsampling as a viable method to reduce resources required for alignment. Further, we propose an information theory-based methodology for efficient alignment by identifying a small high quality subset thereby reducing the computation and time required by alignment. We evaluate the proposed methodology over multiple datasets and compare the results. We find that the model aligned using our proposed methodology outperforms other sampling methods and performs comparable to the model aligned with the full dataset while using less than 10% data, leading to greater than 90% savings in costs, resources, and faster LLM alignment.

Arshika Lalan, Rajat Ghosh, Aditya Kolsur et al.

Recent work explores agentic inference-time techniques to perform structured, multi-step reasoning. However, stateless inference often struggles on multi-step tasks due to the absence of persistent state. Moreover, task-specific fine-tuning or instruction-tuning often achieve surface-level code generation but remain brittle on tasks requiring deeper reasoning and long-horizon dependencies. To address these limitations, we propose stateful multi-agent evolutionary search, a training-free framework that departs from prior stateless approaches by combining (i) persistent inference-time state, (ii) adversarial mutation, and (iii) evolutionary preservation. We demonstrate its effectiveness in automated unit test generation through the generation of edge cases. We generate robust edge cases using an evolutionary search process, where specialized agents sequentially propose, mutate, and score candidates. A controller maintains persistent state across generations, while evolutionary preservation ensures diversity and exploration across all possible cases. This yields a generalist agent capable of discovering robust, high-coverage edge cases across unseen codebases. Experiments show our stateful multi-agent inference framework achieves substantial gains in coverage over stateless single-step baselines, evaluated on prevalent unit-testing benchmarks such as HumanEval and TestGenEvalMini and using three diverse LLM families - Llama, Gemma, and GPT. These results indicate that combining persistent inference-time state with evolutionary search materially improves unit-test generation.

Pratik Shah, Rajat Ghosh, Aryan Singhal et al.

General-purpose automated software engineering (ASE) includes tasks such as code completion, retrieval, repair, QA, and summarization. These tasks require a code retrieval system that can handle specific queries about code entities, or code entity queries (for example, locating a specific class or retrieving the dependencies of a function), as well as general queries without explicit code entities, or natural language queries (for example, describing a task and retrieving the corresponding code). We present RANGER, a repository-level code retrieval agent designed to address both query types, filling a gap in recent works that have focused primarily on code-entity queries. We first present a tool that constructs a comprehensive knowledge graph of the entire repository, capturing hierarchical and cross-file dependencies down to the variable level, and augments graph nodes with textual descriptions and embeddings to bridge the gap between code and natural language. RANGER then operates on this graph through a dual-stage retrieval pipeline. Entity-based queries are answered through fast Cypher lookups, while natural language queries are handled by MCTS-guided graph exploration. We evaluate RANGER across four diverse benchmarks that represent core ASE tasks including code search, question answering, cross-file dependency retrieval, and repository-level code completion. On CodeSearchNet and RepoQA it outperforms retrieval baselines that use embeddings from strong models such as Qwen3-8B. On RepoBench, it achieves superior cross-file dependency retrieval over baselines, and on CrossCodeEval, pairing RANGER with BM25 delivers the highest exact match rate in code completion compared to other RAG methods.

Jinan Zhou, Rajat Ghosh, Vaishnavi Bhargava et al.

When designing LLM services, practitioners care about three key properties: inference-time budget, factual authenticity, and reasoning capacity. However, our analysis shows that no model can simultaneously optimize for all three. We formally prove this trade-off and propose a principled framework named The BAR Theorem for LLM-application design.

Peter Mattson, Christine Cheng, Cody Coleman et al.

Machine learning (ML) needs industry-standard performance benchmarks to support design and competitive evaluation of the many emerging software and hardware solutions for ML. But ML training presents three unique benchmarking challenges absent from other domains: optimizations that improve training throughput can increase the time to solution, training is stochastic and time to solution exhibits high variance, and software and hardware systems are so diverse that fair benchmarking with the same binary, code, and even hyperparameters is difficult. We therefore present MLPerf, an ML benchmark that overcomes these challenges. Our analysis quantitatively evaluates MLPerf's efficacy at driving performance and scalability improvements across two rounds of results from multiple vendors.

Fedor Kitashov, Elizaveta Svitanko, Debojyoti Dutta

State-of-the-art automatic speech recognition (ASR) systems struggle with the lack of data for rare accents. For sufficiently large datasets, neural engines tend to outshine statistical models in most natural language processing problems. However, a speech accent remains a challenge for both approaches. Phonologists manually create general rules describing a speaker's accent, but their results remain underutilized. In this paper, we propose a model that automatically retrieves phonological generalizations from a small dataset. This method leverages the difference in pronunciation between a particular dialect and General American English (GAE) and creates new accented samples of words. The proposed model is able to learn all generalizations that previously were manually obtained by phonologists. We use this statistical method to generate a million phonological variations of words from the CMU Pronouncing Dictionary and train a sequence-to-sequence RNN to recognize accented words with 59% accuracy.

Marc Solanas, Julio Hernandez-Castro, Debojyoti Dutta

More users and companies make use of cloud services every day. They all expect a perfect performance and any issue to remain transparent to them. This last statement is very challenging to perform. A user's activities in our cloud can affect the overall performance of our servers, having an impact on other resources. We can consider these kind of activities as fraudulent. They can be either illegal activities, such as launching a DDoS attack or just activities which are undesired by the cloud provider, such as Bitcoin mining, which uses substantial power, reduces the life of the hardware and can possibly slow down other user's activities. This article discusses a method to detect such activities by using non-intrusive, privacy-friendly data: billing data. We use OpenStack as an example with data provided by Telemetry, the component in charge of measuring resource usage for billing purposes. Results will be shown proving the efficiency of this method and ways to improve it will be provided as well as its advantages and disadvantages.