Aditya Kanade

Lakshya A Agrawal, Aditya Kanade, Navin Goyal et al.

Language models of code (LMs) work well when the surrounding code provides sufficient context. This is not true when it becomes necessary to use types, functionality or APIs defined elsewhere in the repository or a linked library, especially those not seen during training. LMs suffer from limited awareness of such global context and end up hallucinating. Integrated development environments (IDEs) assist developers in understanding repository context using static analysis. We extend this assistance, enjoyed by developers, to LMs. We propose monitor-guided decoding (MGD) where a monitor uses static analysis to guide the decoding. We construct a repository-level dataset PragmaticCode for method-completion in Java and evaluate MGD on it. On models of varying parameter scale, by monitoring for type-consistent object dereferences, MGD consistently improves compilation rates and agreement with ground truth. Further, LMs with fewer parameters, when augmented with MGD, can outperform larger LMs. With MGD, SantaCoder-1.1B achieves better compilation rate and next-identifier match than the much larger text-davinci-003 model. We also conduct a generalizability study to evaluate the ability of MGD to generalize to multiple programming languages (Java, C# and Rust), coding scenarios (e.g., correct number of arguments to method calls), and to enforce richer semantic constraints (e.g., stateful API protocols). Our data and implementation are available at https://github.com/microsoft/monitors4codegen .

Surya Prakash Sahu, Madhurima Mandal, Shikhar Bharadwaj et al. · deepmind

Developers often have questions about semantic aspects of code they are working on, e.g., "Is there a class whose parent classes declare a conflicting attribute?". Answering them requires understanding code semantics such as attributes and inheritance relation of classes. An answer to such a question should identify code spans constituting the answer (e.g., the declaration of the subclass) as well as supporting facts (e.g., the definitions of the conflicting attributes). The existing work on question-answering over code has considered yes/no questions or method-level context. We contribute a labeled dataset, called CodeQueries, of semantic queries over Python code. Compared to the existing datasets, in CodeQueries, the queries are about code semantics, the context is file level and the answers are code spans. We curate the dataset based on queries supported by a widely-used static analysis tool, CodeQL, and include both positive and negative examples, and queries requiring single-hop and multi-hop reasoning. To assess the value of our dataset, we evaluate baseline neural approaches. We study a large language model (GPT3.5-Turbo) in zero-shot and few-shot settings on a subset of CodeQueries. We also evaluate a BERT style model (CuBERT) with fine-tuning. We find that these models achieve limited success on CodeQueries. CodeQueries is thus a challenging dataset to test the ability of neural models, to understand code semantics, in the extractive question-answering setting.

Nalin Wadhwa, Jui Pradhan, Atharv Sonwane et al.

As software projects progress, quality of code assumes paramount importance as it affects reliability, maintainability and security of software. For this reason, static analysis tools are used in developer workflows to flag code quality issues. However, developers need to spend extra efforts to revise their code to improve code quality based on the tool findings. In this work, we investigate the use of (instruction-following) large language models (LLMs) to assist developers in revising code to resolve code quality issues. We present a tool, CORE (short for COde REvisions), architected using a pair of LLMs organized as a duo comprised of a proposer and a ranker. Providers of static analysis tools recommend ways to mitigate the tool warnings and developers follow them to revise their code. The \emph{proposer LLM} of CORE takes the same set of recommendations and applies them to generate candidate code revisions. The candidates which pass the static quality checks are retained. However, the LLM may introduce subtle, unintended functionality changes which may go un-detected by the static analysis. The \emph{ranker LLM} evaluates the changes made by the proposer using a rubric that closely follows the acceptance criteria that a developer would enforce. CORE uses the scores assigned by the ranker LLM to rank the candidate revisions before presenting them to the developer. CORE could revise 59.2% Python files (across 52 quality checks) so that they pass scrutiny by both a tool and a human reviewer. The ranker LLM is able to reduce false positives by 25.8% in these cases. CORE produced revisions that passed the static analysis tool in 76.8% Java files (across 10 quality checks) comparable to 78.3% of a specialized program repair tool, with significantly much less engineering efforts.

Nirmit Arora, Sathvik Joel, Ishan Kavathekar et al.

LLM-based agents are increasingly deployed in multi-agent systems (MAS). As these systems move toward real-world applications, their security becomes paramount. Existing research largely evaluates single-agent security, leaving a critical gap in understanding the vulnerabilities introduced by multi-agent design. However, existing systems fall short due to lack of unified frameworks and metrics focusing on unique rejection modes in MAS. We present SafeAgents, a unified and extensible framework for fine-grained security assessment of MAS. SafeAgents systematically exposes how design choices such as plan construction strategies, inter-agent context sharing, and fallback behaviors affect susceptibility to adversarial prompting. We introduce Dharma, a diagnostic measure that helps identify weak links within multi-agent pipelines. Using SafeAgents, we conduct a comprehensive study across five widely adopted multi-agent architectures (centralized, decentralized, and hybrid variants) on four datasets spanning web tasks, tool use, and code generation. Our findings reveal that common design patterns carry significant vulnerabilities. For example, centralized systems that delegate only atomic instructions to sub-agents obscure harmful objectives, reducing robustness. Our results highlight the need for security-aware design in MAS. Link to code is https://github.com/microsoft/SafeAgents

Prasanna Kumar Routray, Aditya Sanjiv Kanade, Jay Bhanushali et al.

Human texture perception is a weighted average of multi-sensory inputs: visual and tactile. While the visual sensing mechanism extracts global features, the tactile mechanism complements it by extracting local features. The lack of coupled visuotactile datasets in the literature is a challenge for studying multimodal fusion strategies analogous to human texture perception. This paper presents a visual dataset that augments an existing tactile dataset. We propose a novel deep fusion architecture that fuses visual and tactile data using four types of fusion strategies: summation, concatenation, max-pooling, and attention. Our model shows significant performance improvements (97.22%) in surface roughness classification accuracy over tactile only (SVM - 92.60%) and visual only (FENet-50 - 85.01%) architectures. Among the several fusion techniques, attention-guided architecture results in better classification accuracy. Our study shows that analogous to human texture perception, the proposed model chooses a weighted combination of the two modalities (visual and tactile), thus resulting in higher surface roughness classification accuracy; and it chooses to maximize the weightage of the tactile modality where the visual modality fails and vice-versa.

Sai Srinivas Kancheti, Aditya Sanjiv Kanade, Vineeth N. Balasubramanian et al.

Multimodal Reasoning Models (MRMs) leveraging Chain-of-Thought (CoT) based thinking have revolutionized mathematical and logical problem-solving. However, we show that this paradigm struggles with generalized spatial intelligence. We perform a comprehensive evaluation of seventeen models across thirteen spatial benchmarks and identify a critical gap: CoT prompting consistently degrades performance in visual spatial reasoning. Furthermore, through a novel No-Image++ ablation, we demonstrate that MRMs and CoT prompted MLMs suffer from severe shortcut learning, and hallucinate visual details from textual priors even when the image is absent. These findings challenge the efficacy of text-only CoT for spatial tasks and underscore the need for vision-centric reasoning paradigms.

Ramneet Singh, Sathvik Joel, Abhav Mehrotra et al.

Large Language Model (LLM)-based coding agents have shown promising results on coding benchmarks, but their effectiveness on systems code remains underexplored. Due to the size and complexities of systems code, making changes to a systems codebase is a daunting task, even for humans. It requires researching about many pieces of context, derived from the large codebase and its massive commit history, before making changes. Inspired by the recent progress on deep research agents, we design the first deep research agent for code, called Code Researcher, and apply it to the problem of generating patches for mitigating crashes reported in systems code. Code Researcher performs multi-step reasoning about semantics, patterns, and commit history of code to gather sufficient context. The context is stored in a structured memory which is used for synthesizing a patch. We evaluate Code Researcher on kBenchSyz, a benchmark of Linux kernel crashes, and show that it significantly outperforms strong baselines, achieving a crash-resolution rate of 58%, compared to 37.5% by SWE-agent. On an average, Code Researcher explores 10 files in each trajectory whereas SWE-agent explores only 1.33 files, highlighting Code Researcher's ability to deeply explore the codebase. Through another experiment on an open-source multimedia software, we show the generalizability of Code Researcher. Our experiments highlight the importance of global context gathering and multi-faceted reasoning for large codebases.

Aditya Kanade, Mansi Sharma, Manivannan Muniyandi

Physical rehabilitation programs frequently begin with a brief stay in the hospital and continue with home-based rehabilitation. Lack of feedback on exercise correctness is a significant issue in home-based rehabilitation. Automated movement quality assessment (MQA) using skeletal movement data (hereafter referred to as skeletal data) collected via depth imaging devices can assist with home-based rehabilitation by providing the necessary quantitative feedback. This paper aims to use recent advances in deep learning to address the problem of MQA. Movement quality score generation is an essential component of MQA. We propose three novel skeletal data augmentation schemes. We show that using the proposed augmentations for generating movement quality scores result in significant performance boosts over existing methods. Finally, we propose a novel transformer based architecture for MQA. Four novel feature extractors are proposed and studied that allow the transformer network to operate on skeletal data. We show that adding the attention mechanism in the design of the proposed feature extractor allows the transformer network to pay attention to specific body parts that make a significant contribution towards executing a movement. We report an improvement in movement quality score prediction of 12% on UI-PRMD dataset and 21% on KIMORE dataset compared to the existing methods.

Aditya Kanade, Tanuja Ganu

Multimodal Large Language Models (MLLMs) show reasoning promise, yet their visual perception is a critical bottleneck. Strikingly, MLLMs can produce correct answers even while misinterpreting crucial visual elements, masking these underlying failures. Our preliminary study on a joint perception-reasoning dataset revealed that for one leading MLLM, 29% of its correct answers to reasoning questions still exhibited visual perception errors. To systematically address this, we introduce "Do You See Me", a scalable benchmark with 1,758 images and 2,612 questions. It spans seven human-psychology inspired subtasks in 2D and 3D, featuring controllable complexity to rigorously evaluate MLLM visual skills. Our findings on 3 leading closed-source and 5 major open-source models reveal a stark deficit: humans achieve 96.49% accuracy, while top MLLMs average below 50%. This performance gap widens rapidly with increased task complexity (e.g., from 12% to 45% in the visual form constancy subtask). Further analysis into the root causes suggests that failures stem from challenges like misallocated visual attention and the instability of internal representations for fine-grained details, especially at or below encoder patch resolution. This underscores an urgent need for MLLMs with truly robust visual perception. The benchmark dataset, source code and evaluation scripts are available at https://github.com/microsoft/Do-You-See-Me.

Mohammad Saifullah, Thomas Kornmaier, Taaha Kazi et al.

Long-form fiction writers need memory that answers multi-hop questions about evolving story state: who knows a secret and when they learned it, whether an event preceded the narration that revealed it, whether a setup paid off, and how a relationship shifted. General-purpose retrieval and agent-memory systems represent entities and facts but not the narratological structure these questions turn on, so they surface the wrong evidence or none at all. We introduce the Narrative World Model (NWM), a writer-memory system that pairs a narratology-grounded typed temporal-state graph with query-conditioned hybrid retrieval. To measure memory rather than the answerer, we read every system through a single held-constant Opus 4.8 reader over only that system's chapter-safe evidence, on a reproducible public corpus and a validated multi-hop benchmark, and we compare against the strongest existing temporal-knowledge-graph agent-memory framework, Graphiti/Zep (Rasmussen et al., 2025). NWM substantially and significantly outperforms this baseline on multi-hop narratological QA across both corpora, and far exceeds GraphRAG and flat retrieval. The advantage is representational rather than an artifact of extraction: it survives rebuilding the baseline with NWM's own extractor, and traces to its narratology-grounded structure and query-conditioned retrieval, not to graph size or extractor quality.

Aditya Kanade, Petros Maniatis, Gogul Balakrishnan et al.

Recent research has achieved impressive results on understanding and improving source code by building up on machine-learning techniques developed for natural languages. A significant advancement in natural-language understanding has come with the development of pre-trained contextual embeddings, such as BERT, which can be fine-tuned for downstream tasks with less labeled data and training budget, while achieving better accuracies. However, there is no attempt yet to obtain a high-quality contextual embedding of source code, and to evaluate it on multiple program-understanding tasks simultaneously; that is the gap that this paper aims to mitigate. Specifically, first, we curate a massive, deduplicated corpus of 7.4M Python files from GitHub, which we use to pre-train CuBERT, an open-sourced code-understanding BERT model; and, second, we create an open-sourced benchmark that comprises five classification tasks and one program-repair task, akin to code-understanding tasks proposed in the literature before. We fine-tune CuBERT on our benchmark tasks, and compare the resulting models to different variants of Word2Vec token embeddings, BiLSTM and Transformer models, as well as published state-of-the-art models, showing that CuBERT outperforms them all, even with shorter training, and with fewer labeled examples. Future work on source-code embedding can benefit from reusing our benchmark, and from comparing against CuBERT models as a strong baseline.

Manav Singhal, Tushar Aggarwal, Abhijeet Awasthi et al.

Existing evaluation benchmarks of language models of code (code LMs) focus almost exclusively on whether the LMs can generate functionally-correct code. In real-world software engineering, developers think beyond functional correctness. They have requirements on "how" a functionality should be implemented to meet overall system design objectives like efficiency, security, and maintainability. They would also trust the code LMs more if the LMs demonstrate robust understanding of such requirements. We propose a new benchmark NoFunEval to evaluate code LMs on non-functional requirements and simple classification instances for both functional and non-functional requirements. We propose a prompting method, Coding Concepts (CoCo), as a way for a developer to communicate the domain knowledge to the LMs. We conduct an extensive evaluation of 27 code LMs. Our finding is that LMs generally falter when tested on our benchmark, hinting at fundamental blindspots in their training setups. Surprisingly, even the classification accuracy on functional-correctness instances derived from the popular HumanEval benchmark is low, calling in question the depth of their comprehension and the source of their success in generating functionally-correct code in the first place. We release our benchmark and evaluation scripts publicly at https://aka.ms/NoFunEval.

Ajinkya Deshpande, Anmol Agarwal, Shashank Shet et al.

LLMs have demonstrated significant potential in code generation tasks, achieving promising results at the function or statement level across various benchmarks. However, the complexities associated with creating code artifacts like classes, particularly within the context of real-world software repositories, remain underexplored. Prior research treats class-level generation as an isolated task, neglecting the intricate dependencies & interactions that characterize real-world software environments. To address this gap, we introduce RepoClassBench, a comprehensive benchmark designed to rigorously evaluate LLMs in generating complex, class-level code within real-world repositories. RepoClassBench includes "Natural Language to Class generation" tasks across Java, Python & C# from a selection of repositories. We ensure that each class in our dataset not only has cross-file dependencies within the repository but also includes corresponding test cases to verify its functionality. We find that current models struggle with the realistic challenges posed by our benchmark, primarily due to their limited exposure to relevant repository contexts. To address this shortcoming, we introduce Retrieve-Repotools-Reflect (RRR), a novel approach that equips LLMs with static analysis tools to iteratively navigate & reason about repository-level context in an agent-based framework. Our experiments demonstrate that RRR significantly outperforms existing baselines on RepoClassBench, showcasing its effectiveness across programming languages & under various settings. Our findings emphasize the critical need for code-generation benchmarks to incorporate repo-level dependencies to more accurately reflect the complexities of software development. Our work shows the benefits of leveraging specialized tools to enhance LLMs' understanding of repository context. We plan to make our dataset & evaluation harness public.

Aniketh Garikaparthi, Manasi Patwardhan, Aditya Sanjiv Kanade et al.

There has been a surge of interest in harnessing the reasoning capabilities of Large Language Models (LLMs) to accelerate scientific discovery. While existing approaches rely on grounding the discovery process within the relevant literature, effectiveness varies significantly with the quality and nature of the retrieved literature. We address the challenge of retrieving prior work whose concepts can inspire solutions for a given research problem, a task we define as Methodology Inspiration Retrieval (MIR). We construct a novel dataset tailored for training and evaluating retrievers on MIR, and establish baselines. To address MIR, we build the Methodology Adjacency Graph (MAG); capturing methodological lineage through citation relationships. We leverage MAG to embed an "intuitive prior" into dense retrievers for identifying patterns of methodological inspiration beyond superficial semantic similarity. This achieves significant gains of +5.4 in Recall@3 and +7.8 in Mean Average Precision (mAP) over strong baselines. Further, we adapt LLM-based re-ranking strategies to MIR, yielding additional improvements of +4.5 in Recall@3 and +4.8 in mAP. Through extensive ablation studies and qualitative analyses, we exhibit the promise of MIR in enhancing automated scientific discovery and outline avenues for advancing inspiration-driven retrieval.

Priyanshu Gupta, Avishree Khare, Yasharth Bajpai et al.

Developers expend a significant amount of time in editing code for a variety of reasons such as bug fixing or adding new features. Designing effective methods to predict code edits has been an active yet challenging area of research due to the diversity of code edits and the difficulty of capturing the developer intent. In this work, we address these challenges by endowing pre-trained large language models (LLMs) of code with the knowledge of prior, relevant edits. The generative capability of the LLMs helps address the diversity in code changes and conditioning code generation on prior edits helps capture the latent developer intent. We evaluate two well-known LLMs, Codex and CodeT5, in zero-shot and fine-tuning settings respectively. In our experiments with two datasets, the knowledge of prior edits boosts the performance of the LLMs significantly and enables them to generate 29% and 54% more correctly edited code in top-1 suggestions relative to the current state-of-the-art symbolic and neural approaches, respectively.

Aditya Kanade, Mansi Sharma, M. Manivannan

Technology has an important role to play in the field of Rehabilitation, improving patient outcomes and reducing healthcare costs. However, existing approaches lack clinical validation, robustness and ease of use. We propose Tele-EvalNet, a novel system consisting of two components: a live feedback model and an overall performance evaluation model. The live feedback model demonstrates feedback on exercise correctness with easy to understand instructions highlighted using color markers. The overall performance evaluation model learns a mapping of joint data to scores, given to the performance by clinicians. The model does this by extracting clinically approved features from joint data. Further, these features are encoded to a lower dimensional space with an autoencoder. A novel multi-scale CNN-LSTM network is proposed to learn a mapping of performance data to the scores by leveraging features extracted at multiple scales. The proposed system shows a high degree of improvement in score predictions and outperforms the state-of-the-art rehabilitation models.