Naman Jain

Naman Jain, Tianjun Zhang, Wei-Lin Chiang et al. · berkeley, microsoft-research

Natural language to code generation is an important application area of LLMs and has received wide attention from the community. The majority of relevant studies have exclusively concentrated on increasing the quantity and functional correctness of training sets while disregarding other stylistic elements of programs. More recently, data quality has garnered a lot of interest and multiple works have showcased its importance for improving performance. In this work, we investigate data quality for code and find that making the code more structured and readable leads to improved code generation performance of the system. We build a novel data-cleaning pipeline that uses these principles to transform existing programs by 1.) renaming variables, 2.) modularizing and decomposing complex code into smaller helper sub-functions, and 3.) inserting natural-language based plans via LLM based transformations. We evaluate our approach on two challenging algorithmic code generation benchmarks and find that fine-tuning CodeLLaMa-7B on our transformed modularized programs improves the performance by up to 30% compared to fine-tuning on the original dataset. Additionally, we demonstrate improved performance from using a smaller amount of higher-quality data, finding that a model fine-tuned on the entire original dataset is outperformed by a model trained on 15% of our cleaned dataset. Even in comparison to closed-source models, our models outperform the much larger AlphaCoder models.

Cursor Research, Aaron Chan, Ahmed Shalaby et al. · berkeley, microsoft-research

Composer 2 is a specialized model designed for agentic software engineering. The model demonstrates strong long-term planning and coding intelligence while maintaining the ability to efficiently solve problems for interactive use. The model is trained in two phases: first, continued pretraining to improve the model's knowledge and latent coding ability, followed by large-scale reinforcement learning to improve end-to-end coding performance through stronger reasoning, accurate multi-step execution, and coherence on long-horizon realistic coding problems. We develop infrastructure to support training in the same Cursor harness that is used by the deployed model, with equivalent tools and structure, and use environments that match real problems closely. To measure the ability of the model on increasingly difficult tasks, we introduce a benchmark derived from real software engineering problems in large codebases including our own. Composer 2 is a frontier-level coding model and demonstrates a process for training strong domain-specialized models. On our CursorBench evaluations the model achieves a major improvement in accuracy compared to previous Composer models (61.3). On public benchmarks the model scores 61.7 on Terminal-Bench and 73.7 on SWE-bench Multilingual in our harness, comparable to state-of-the-art systems.

Aditya G. Parameswaran, Shreya Shankar, Parth Asawa et al.

Large language models (LLMs) are incredibly powerful at comprehending and generating data in the form of text, but are brittle and error-prone. There has been an advent of toolkits and recipes centered around so-called prompt engineering-the process of asking an LLM to do something via a series of prompts. However, for LLM-powered data processing workflows, in particular, optimizing for quality, while keeping cost bounded, is a tedious, manual process. We put forth a vision for declarative prompt engineering. We view LLMs like crowd workers and leverage ideas from the declarative crowdsourcing literature-including leveraging multiple prompting strategies, ensuring internal consistency, and exploring hybrid-LLM-non-LLM approaches-to make prompt engineering a more principled process. Preliminary case studies on sorting, entity resolution, and imputation demonstrate the promise of our approach

Mohit Jaiswal, Naman Jain, Shivani Pathak et al.

Few-shot remote sensing image classification is challenging due to limited labeled samples and high variability in land-cover types. We propose a reconstruction-guided few-shot network (RGFS-Net) that enhances generalization to unseen classes while preserving consistency for seen categories. Our method incorporates a masked image reconstruction task, where parts of the input are occluded and reconstructed to encourage semantically rich feature learning. This auxiliary task strengthens spatial understanding and improves class discrimination under low-data settings. We evaluated the efficacy of EuroSAT and PatternNet datasets under 1-shot and 5-shot protocols, our approach consistently outperforms existing baselines. The proposed method is simple, effective, and compatible with standard backbones, offering a robust solution for few-shot remote sensing classification. Codes are available at https://github.com/stark0908/RGFS.

Tianjun Zhang, Shishir G. Patil, Naman Jain et al. · berkeley, microsoft-research

Pretraining Large Language Models (LLMs) on large corpora of textual data is now a standard paradigm. When using these LLMs for many downstream applications, it is common to additionally bake in new knowledge (e.g., time-critical news, or private domain knowledge) into the pretrained model either through RAG-based-prompting, or fine-tuning. However, the optimal methodology for the model to gain such new knowledge remains an open question. In this paper, we present Retrieval Augmented FineTuning (RAFT), a training recipe that improves the model's ability to answer questions in a "open-book" in-domain settings. In RAFT, given a question, and a set of retrieved documents, we train the model to ignore those documents that don't help in answering the question, which we call, distractor documents. RAFT accomplishes this by citing verbatim the right sequence from the relevant document that would help answer the question. This coupled with RAFT's chain-of-thought-style response helps improve the model's ability to reason. In domain-specific RAG, RAFT consistently improves the model's performance across PubMed, HotpotQA, and Gorilla datasets, presenting a post-training recipe to improve pre-trained LLMs to in-domain RAG. RAFT's code and demo are open-sourced at github.com/ShishirPatil/gorilla.

Naman Jain, Jaskirat Singh, Manish Shetty et al. · berkeley, microsoft-research

Improving open-source models on real-world SWE tasks (solving GITHUB issues) faces two key challenges: 1) scalable curation of execution environments to train these models, and, 2) optimal scaling of test-time compute. We introduce AgentGym, the largest procedurally-curated executable gym environment for training real-world SWE-agents, consisting of more than 8.7K tasks. AgentGym is powered by two main contributions: 1) SYNGEN: a synthetic data curation recipe that enables scalable curation of executable environments using test-generation and back-translation directly from commits, thereby reducing reliance on human-written issues or unit tests. We show that this enables more scalable training leading to pass@1 performance of 34.4% on SWE-Bench Verified benchmark with our 32B model. 2) Hybrid Test-time Scaling: we provide an in-depth analysis of two test-time scaling axes; execution-based and execution-free verifiers, demonstrating that they exhibit complementary strengths and limitations. Test-based verifiers suffer from low distinguishability, while execution-free verifiers are biased and often rely on stylistic features. Surprisingly, we find that while each approach individually saturates around 42-43%, significantly higher gains can be obtained by leveraging their complementary strengths. Overall, our approach achieves 51% on the SWE-Bench Verified benchmark, reflecting a new state-of-the-art for open-weight SWE-agents and for the first time showing competitive performance with proprietary models such as o1, o1-preview and sonnet-3.5-v2 (with tools). We will open-source our environments, models, and agent trajectories.

Naman Jain, King Han, Alex Gu et al. · berkeley, microsoft-research

Large Language Models (LLMs) applied to code-related applications have emerged as a prominent field, attracting significant interest from both academia and industry. However, as new and improved LLMs are developed, existing evaluation benchmarks (e.g., HumanEval, MBPP) are no longer sufficient for assessing their capabilities. In this work, we propose LiveCodeBench, a comprehensive and contamination-free evaluation of LLMs for code, which continuously collects new problems over time from contests across three competition platforms, namely LeetCode, AtCoder, and CodeForces. Notably, our benchmark also focuses on a broader range of code related capabilities, such as self-repair, code execution, and test output prediction, beyond just code generation. Currently, LiveCodeBench hosts four hundred high-quality coding problems that were published between May 2023 and May 2024. We have evaluated 18 base LLMs and 34 instruction-tuned LLMs on LiveCodeBench. We present empirical findings on contamination, holistic performance comparisons, potential overfitting in existing benchmarks as well as individual model comparisons. We will release all prompts and model completions for further community analysis, along with a general toolkit for adding new scenarios and model

Anton Lozhkov, Raymond Li, Loubna Ben Allal et al. · berkeley, ibm-research

The BigCode project, an open-scientific collaboration focused on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder2. In partnership with Software Heritage (SWH), we build The Stack v2 on top of the digital commons of their source code archive. Alongside the SWH repositories spanning 619 programming languages, we carefully select other high-quality data sources, such as GitHub pull requests, Kaggle notebooks, and code documentation. This results in a training set that is 4x larger than the first StarCoder dataset. We train StarCoder2 models with 3B, 7B, and 15B parameters on 3.3 to 4.3 trillion tokens and thoroughly evaluate them on a comprehensive set of Code LLM benchmarks. We find that our small model, StarCoder2-3B, outperforms other Code LLMs of similar size on most benchmarks, and also outperforms StarCoderBase-15B. Our large model, StarCoder2- 15B, significantly outperforms other models of comparable size. In addition, it matches or outperforms CodeLlama-34B, a model more than twice its size. Although DeepSeekCoder- 33B is the best-performing model at code completion for high-resource languages, we find that StarCoder2-15B outperforms it on math and code reasoning benchmarks, as well as several low-resource languages. We make the model weights available under an OpenRAIL license and ensure full transparency regarding the training data by releasing the SoftWare Heritage persistent IDentifiers (SWHIDs) of the source code data.

Naman Jain, Ankush Chauhan, Atharva Chewale et al.

Song lyrics convey a meaningful story in a creative manner with complex rhythmic patterns. Researchers have been successful in generating and analyisng lyrics for poetry and songs in English and Chinese. But there are no works which explore the Hindi language datasets. Given the popularity of Hindi songs across the world and the ambiguous nature of romanized Hindi script, we propose Bollyrics, an automatic lyric generator for romanized Hindi songs. We propose simple techniques to capture rhyming patterns before and during the model training process in Hindi language. The dataset and codes are available publicly at https://github.com/lingo-iitgn/Bollyrics.

Yuxiang Wei, Federico Cassano, Jiawei Liu et al.

Instruction tuning is a supervised fine-tuning approach that significantly improves the ability of large language models (LLMs) to follow human instructions. We propose SelfCodeAlign, the first fully transparent and permissive pipeline for self-aligning code LLMs without extensive human annotations or distillation. SelfCodeAlign employs the same base model for inference throughout the data generation process. It first extracts diverse coding concepts from high-quality seed snippets to generate new tasks. It then samples multiple responses per task, pairs each with test cases, and validates them in a sandbox environment. Finally, passing examples are selected for instruction tuning. In our primary experiments, we use SelfCodeAlign with CodeQwen1.5-7B to generate a dataset of 74k instruction-response pairs. Finetuning on this dataset leads to a model that achieves a 67.1 pass@1 on HumanEval+, surpassing CodeLlama-70B-Instruct despite being ten times smaller. Across all benchmarks, this finetuned model consistently outperforms the original version trained with OctoPack, the previous state-of-the-art method for instruction tuning without human annotations or distillation. Additionally, we show that SelfCodeAlign is effective across LLMs of various sizes, from 3B to 33B, and that the base models can benefit more from alignment with their own data distribution. We further validate each component's effectiveness in our pipeline, showing that SelfCodeAlign outperforms both direct distillation from GPT-4o and leading GPT-3.5-based distillation methods, such as OSS-Instruct and Evol-Instruct. SelfCodeAlign has also led to the creation of StarCoder2-Instruct, the first fully transparent, permissively licensed, and self-aligned code LLM that achieves state-of-the-art coding performance.

Naman Jain, Pranjali Jain, Pratik Kayal et al.

India is an agro-based economy and proper information about agricultural practices is the key to optimal agricultural growth and output. In order to answer the queries of the farmer, we have build an agricultural chatbot based on the dataset from Kisan Call Center. This system is robust enough to answer queries related to weather, market rates, plant protection and government schemes. This system is available 24* 7, can be accessed through any electronic device and the information is delivered with the ease of understanding. The system is based on a sentence embedding model which gives an accuracy of 56%. After eliminating synonyms and incorporating entity extraction, the accuracy jumps to 86%. With such a system, farmers can progress towards easier information about farming related practices and hence a better agricultural output. The job of the Call Center workforce would be made easier and the hard work of various such workers can be redirected to a better goal.

Alex Gu, Wen-Ding Li, Naman Jain et al. · berkeley, microsoft-research

While language models are increasingly more proficient at code generation, they still frequently generate incorrect programs. Many of these programs are obviously wrong, but others are more subtle and pass weaker correctness checks such as being able to compile. In this work, we focus on these counterfeit samples: programs sampled from a language model that 1) have a high enough log-probability to be generated at a moderate temperature and 2) pass weak correctness checks. Overall, we discover that most models have a very shallow understanding of counterfeits through three clear failure modes. First, models mistakenly classify them as correct. Second, models are worse at reasoning about the execution behaviour of counterfeits and often predict their execution results as if they were correct. Third, when asking models to fix counterfeits, the likelihood of a model successfully repairing a counterfeit is often even lower than that of sampling a correct program from scratch. Counterfeits also have very unexpected properties: first, counterfeit programs for problems that are easier for a model to solve are not necessarily easier to detect and only slightly easier to execute and repair. Second, counterfeits from a given model are just as confusing to the model itself as they are to other models. Finally, both strong and weak models are able to generate counterfeit samples that equally challenge all models. In light of our findings, we recommend that care and caution be taken when relying on models to understand their own samples, especially when no external feedback is incorporated.

Manish Shetty, Naman Jain, Adwait Godbole et al. · berkeley, microsoft-research

Despite extensive usage in high-performance, low-level systems programming applications, C is susceptible to vulnerabilities due to manual memory management and unsafe pointer operations. Rust, a modern systems programming language, offers a compelling alternative. Its unique ownership model and type system ensure memory safety without sacrificing performance. In this paper, we present Syzygy, an automated approach to translate C to safe Rust. Our technique uses a synergistic combination of LLM-driven code and test translation guided by dynamic-analysis-generated execution information. This paired translation runs incrementally in a loop over the program in dependency order of the code elements while maintaining per-step correctness. Our approach exposes novel insights on combining the strengths of LLMs and dynamic analysis in the context of scaling and combining code generation with testing. We apply our approach to successfully translate Zopfli, a high-performance compression library with ~3000 lines of code and 98 functions. We validate the translation by testing equivalence with the source C program on a set of inputs. To our knowledge, this is the largest automated and test-validated C to safe Rust code translation achieved so far.

Manish Shetty, Naman Jain, Jinjian Liu et al. · berkeley, microsoft-research

Developing high-performance software is a complex task that requires specialized expertise. We introduce GSO, a benchmark for evaluating language models' capabilities in developing high-performance software. We develop an automated pipeline that generates and executes performance tests to analyze repository commit histories to identify 102 challenging optimization tasks across 10 codebases, spanning diverse domains and programming languages. An agent is provided with a codebase and performance test as a precise specification, and tasked to improve the runtime efficiency, which is measured against the expert developer optimization. Our quantitative evaluation reveals that leading SWE-Agents struggle significantly, achieving less than 5% success rate, with limited improvements even with inference-time scaling. Our qualitative analysis identifies key failure modes, including difficulties with low-level languages, practicing lazy optimization strategies, and challenges in accurately localizing bottlenecks. We release the code and artifacts of our benchmark along with agent trajectories to enable future research.

Alex Gu, Naman Jain, Wen-Ding Li et al. · berkeley, microsoft-research

AI for software engineering has made remarkable progress recently, becoming a notable success within generative AI. Despite this, there are still many challenges that need to be addressed before automated software engineering reaches its full potential. It should be possible to reach high levels of automation where humans can focus on the critical decisions of what to build and how to balance difficult tradeoffs while most routine development effort is automated away. Reaching this level of automation will require substantial research and engineering efforts across academia and industry. In this paper, we aim to discuss progress towards this in a threefold manner. First, we provide a structured taxonomy of concrete tasks in AI for software engineering, emphasizing the many other tasks in software engineering beyond code generation and completion. Second, we outline several key bottlenecks that limit current approaches. Finally, we provide an opinionated list of promising research directions toward making progress on these bottlenecks, hoping to inspire future research in this rapidly maturing field.

Naman Jain, Amir Kalev

We introduce Quantum Feature Extraction (QuFeX), a novel quantum machine learning module. The proposed module enables feature extraction in a reduced-dimensional space, significantly decreasing the number of parallel evaluations required in typical quantum convolutional neural network architectures. Its design allows seamless integration into deep classical neural networks, making it particularly suitable for hybrid quantum-classical models. As an application of QuFeX, we propose Qu-Net -- a hybrid architecture which integrates QuFeX at the bottleneck of a U-Net architecture. The latter is widely used for image segmentation tasks such as medical imaging and autonomous driving. Our numerical analysis indicates that the Qu-Net can achieve superior segmentation performance compared to a U-Net baseline. These results highlight the potential of QuFeX to enhance deep neural networks by leveraging hybrid computational paradigms, providing a path towards a robust framework for real-world applications requiring precise feature extraction.

Terry Yue Zhuo, Minh Chien Vu, Jenny Chim et al.

Task automation has been greatly empowered by the recent advances in Large Language Models (LLMs) via Python code, where the tasks ranging from software engineering development to general-purpose reasoning. While current benchmarks have shown that LLMs can solve tasks using programs like human developers, the majority of their evaluations are limited to short and self-contained algorithmic tasks or standalone function calls. Solving challenging and practical tasks requires the capability of utilizing diverse function calls as tools to efficiently implement functionalities like data analysis and web development. In addition, using multiple tools to solve a task needs compositional reasoning by accurately understanding complex instructions. Fulfilling both of these characteristics can pose a great challenge for LLMs.To assess how well LLMs can solve challenging and practical tasks via programs, we introduce BigCodeBench, a benchmark that challenges LLMs to invoke multiple function calls as tools from 139 libraries and 7 domains for 1,140 fine-grained tasks. To evaluate LLMs rigorously, each task encompasses 5.6 test cases with an average branch coverage of 99%. In addition, we propose a natural-language-oriented variant of BigCodeBench, BigCodeBench-Instruct, that automatically transforms the original docstrings into short instructions only with essential information. Our extensive evaluation of 60 LLMs shows that LLMs are not yet capable of following complex instructions to use function calls precisely, with scores up to 60%, significantly lower than the human performance of 97%. The results underscore the need for further advancements in this area.

Naman Jain, Skanda Vaidyanath, Arun Iyer et al.

Large pre-trained language models such as GPT-3, Codex, and Google's language model are now capable of generating code from natural language specifications of programmer intent. We view these developments with a mixture of optimism and caution. On the optimistic side, such large language models have the potential to improve productivity by providing an automated AI pair programmer for every programmer in the world. On the cautionary side, since these large language models do not understand program semantics, they offer no guarantees about quality of the suggested code. In this paper, we present an approach to augment these large language models with post-processing steps based on program analysis and synthesis techniques, that understand the syntax and semantics of programs. Further, we show that such techniques can make use of user feedback and improve with usage. We present our experiences from building and evaluating such a tool jigsaw, targeted at synthesizing code for using Python Pandas API using multi-modal inputs. Our experience suggests that as these large language models evolve for synthesizing code from intent, jigsaw has an important role to play in improving the accuracy of the systems.

Akhilesh Ravi, Amit Yadav, Jainish Chauhan et al.

The increasing use of dialogue agents makes it extremely desirable for them to understand and acknowledge the implied emotions to respond like humans with empathy. Chatbots using traditional techniques analyze emotions based on the context and meaning of the text and lack the understanding of emotions expressed through face. Emojis representing facial expressions present a promising way to express emotions. However, none of the AI systems utilizes emojis for empathetic conversation generation. We propose, SentEmojiBot, based on the SentEmoji dataset, to generate empathetic conversations with a combination of emojis and text. Evaluation metrics show that the BERT-based model outperforms the vanilla transformer model. A user study indicates that the dialogues generated by our model were understandable and adding emojis improved empathetic traits in conversations by 9.8%

Ajaykrishna Karthikeyan, Naman Jain, Nagarajan Natarajan et al.

Decision trees provide a rich family of highly non-linear but efficient models, due to which they continue to be the go-to family of predictive models by practitioners across domains. But learning trees is challenging due to their discrete decision boundaries. The state-of-the-art (SOTA) techniques resort to (a) learning \textit{soft} trees thereby losing logarithmic inference time; or (b) using methods tailored to specific supervised learning settings, requiring access to labeled examples and loss function. In this work, by leveraging techniques like overparameterization and straight-through estimators, we propose a unified method that enables accurate end-to-end gradient based tree training and can be deployed in a variety of settings like offline supervised learning and online learning with bandit feedback. Using extensive validation on standard benchmarks, we demonstrate that our method provides best of both worlds, i.e., it is competitive to, and in some cases more accurate than methods designed \textit{specifically} for the supervised settings; and in bandit settings, where most existing tree learning techniques are not applicable, our models are still accurate and significantly outperform the applicable SOTA methods.

Sriram Balasubramanian, Naman Jain, Gaurav Jindal et al.

We evaluate named entity representations of BERT-based NLP models by investigating their robustness to replacements from the same typed class in the input. We highlight that on several tasks while such perturbations are natural, state of the art trained models are surprisingly brittle. The brittleness continues even with the recent entity-aware BERT models. We also try to discern the cause of this non-robustness, considering factors such as tokenization and frequency of occurrence. Then we provide a simple method that ensembles predictions from multiple replacements while jointly modeling the uncertainty of type annotations and label predictions. Experiments on three NLP tasks show that our method enhances robustness and increases accuracy on both natural and adversarial datasets.

Davinder Singh, Naman Jain, Pranjali Jain et al.

India loses 35% of the annual crop yield due to plant diseases. Early detection of plant diseases remains difficult due to the lack of lab infrastructure and expertise. In this paper, we explore the possibility of computer vision approaches for scalable and early plant disease detection. The lack of availability of sufficiently large-scale non-lab data set remains a major challenge for enabling vision based plant disease detection. Against this background, we present PlantDoc: a dataset for visual plant disease detection. Our dataset contains 2,598 data points in total across 13 plant species and up to 17 classes of diseases, involving approximately 300 human hours of effort in annotating internet scraped images. To show the efficacy of our dataset, we learn 3 models for the task of plant disease classification. Our results show that modelling using our dataset can increase the classification accuracy by up to 31%. We believe that our dataset can help reduce the entry barrier of computer vision techniques in plant disease detection.

Monarch Parmar, Naman Jain, Pranjali Jain et al.

Understanding the current research trends, problems, and their innovative solutions remains a bottleneck due to the ever-increasing volume of scientific articles. In this paper, we propose NLPExplorer, a completely automatic portal for indexing, searching, and visualizing Natural Language Processing (NLP) research volume. NLPExplorer presents interesting insights from papers, authors, venues, and topics. In contrast to previous topic modelling based approaches, we manually curate five course-grained non-exclusive topical categories namely Linguistic Target (Syntax, Discourse, etc.), Tasks (Tagging, Summarization, etc.), Approaches (unsupervised, supervised, etc.), Languages (English, Chinese,etc.) and Dataset types (news, clinical notes, etc.). Some of the novel features include a list of young popular authors, popular URLs, and datasets, a list of topically diverse papers and recent popular papers. Also, it provides temporal statistics such as yearwise popularity of topics, datasets, and seminal papers. To facilitate future research and system development, we make all the processed datasets accessible through API calls. The current system is available at http://lingo.iitgn.ac.in:5001/

Sahil Shah, Naman Jain, Abhishek Sharma et al.

This paper provides a comprehensive and exhaustive study of adversarial attacks on human pose estimation models and the evaluation of their robustness. Besides highlighting the important differences between well-studied classification and human pose-estimation systems w.r.t. adversarial attacks, we also provide deep insights into the design choices of pose-estimation systems to shape future work. We benchmark the robustness of several 2D single person pose-estimation architectures trained on multiple datasets, MPII and COCO. In doing so, we also explore the problem of attacking non-classification networks including regression based networks, which has been virtually unexplored in the past. \par We find that compared to classification and semantic segmentation, human pose estimation architectures are relatively robust to adversarial attacks with the single-step attacks being surprisingly ineffective. Our study shows that the heatmap-based pose-estimation models are notably robust than their direct regression-based systems and that the systems which explicitly model anthropomorphic semantics of human body fare better than their other counterparts. Besides, targeted attacks are more difficult to obtain than un-targeted ones and some body-joints are easier to fool than the others. We present visualizations of universal perturbations to facilitate unprecedented insights into their workings on pose-estimation. Additionally, we show them to generalize well across different networks. Finally we perform a user study about perceptibility of these examples.