arXiv subjects

Mayank Agarwal

Publications and source records attributed to Mayank Agarwal.

At least 19 recordsLinked to original sources

arXiv2026

How Good Are LLMs at Processing Tool Outputs?

Most realistic task automation problems require large language models (LLMs) to call tools, which often return complex JSON responses. These responses must be further processed to derive the information necessary for task completion. The ability of LLMs to do so is under-studied. In this paper, we study the tool response processing task and LLMs' abilities to process structured (JSON) responses. We created a dataset for this task, and evaluated 15 open and closed weight models using multiple prompting approaches. Our results show that JSON processing remains a difficult task even for frontier models across multiple prompting strategies. The optimal response processing strategy depends on both the nature and size of the tool outputs, as well as the complexity of the required reasoning. Variations in processing approaches can lead to performance differences ranging from 3\% to 50\%.

cs.LG↗

arXiv2026

ToolRM: Outcome Reward Models for Tool-Calling Large Language Models

As large language models (LLMs) increasingly interact with external tools, reward modeling for tool use has emerged as a critical yet underexplored area of research. Existing reward models, trained primarily on natural language outputs, struggle to evaluate tool-based reasoning and execution. To quantify this gap, we introduce FC-RewardBench, the first benchmark to systematically evaluate reward models in tool-calling scenarios. Our analysis shows that current reward models frequently miss key signals of effective tool use, highlighting the need for domain-specific modeling. We address this by proposing a training framework for outcome reward models using data synthesized from permissively licensed, open-weight LLMs. We introduce ToolRM - a suite of reward models for tool-use ranging from 1.7B to 14B parameters. Across diverse settings, these models consistently outperform general-purpose baselines. Notably, they achieve up to a 25% improvement with Best-of-N sampling, while also improving robustness to input noise, enabling effective data filtering, and supporting RL-training of policy models.

cs.CL↗

arXiv2025

Assessing Engineering Student Perceptions of Introductory CS Courses in an Indian Context

Understanding student perceptions of assessment is vital for designing inclusive and effective learning environments, especially in technical education. This study explores engineering students' perceptions of assessment practices in an introductory computer science/ programming course, and its associated laboratory within an Indian engineering institute context. A total of 318 first-year Bachelor of Technology students participated in a weekly 25-statement Likert-scale survey conducted over nine weeks. Using descriptive statistics and non-parametric tests (Mann-Whitney U and Kruskal-Wallis), the analysis reveals that students largely perceive lab assignments as effective learning activities and view exams and projects as authentic and skill-enhancing. Students appreciated the role of instructors in shaping course content and found teaching assistants to be approachable and helpful, despite some inconsistencies. The study also finds significant variations in students' academic performance and assessment perceptions based on prior programming experience, technology familiarity, gender, and academic branch. Notably, the performance data did not follow a Gaussian distribution, challenging common assumptions in grade modeling. A comparative analysis with European cohorts highlights both universal patterns and contextual differences, offering valuable insights for designing inclusive and equitable assessment strategies in programming education.

cs.CY↗

arXiv2025

NESTFUL: A Benchmark for Evaluating LLMs on Nested Sequences of API Calls

The resurgence of autonomous agents built using large language models (LLMs) to solve complex real-world tasks has brought increased focus on LLMs' fundamental ability of tool or function calling. At the core of these agents, an LLM must plan, execute, and respond using external tools, APIs, and custom functions. Research on tool calling has gathered momentum, but evaluation benchmarks and datasets representing the complexity of the tasks have lagged behind. In this work, we focus on one such complexity, nested sequencing, with the goal of extending existing benchmarks and evaluation. Specifically, we present NESTFUL, a benchmark to evaluate LLMs on nested sequences of API calls, i.e., sequences where the output of one API call is passed as input to a subsequent call. NESTFUL contains 1800+ nested sequences where all the function calls are executable. Experimental results on a variety of models show that the best-performing model (GPT-4o) achieves a full sequence match accuracy of 28% and a win-rate of 60%, necessitating a large scope for improvement in the nested sequencing aspect of function calling. Our analysis of these results provides possible future research directions for the community, in addition to a benchmark to track progress. We have released the NESTFUL dataset under the Apache 2.0 license at https://github.com/IBM/NESTFUL.

cs.AI↗

arXiv2025

LongFuncEval: Measuring the effectiveness of long context models for function calling

Multiple recent studies have documented large language models' (LLMs) performance on calling external tools/functions. Others focused on LLMs' abilities to handle longer context lengths. At the intersection of these areas lies another interesting problem: LLMs' abilities to accurately perform function calls in long context settings. Particularly, when calling tools, LLMs are encumbered by three predominant challenges: (1) a large catalog of tools, (2) long responses from the tool APIs, and (3) long multi-turn conversations. These challenges are particularly relevant to enterprise applications of LLMs which engage in multi-turn conversations with users to complete complex tasks that require a large catalog of complex tools. The literature contains multiple investigations of long context challenges such as lost in the middle or needle in the haystack for natural language tasks. In this paper, we make the first attempt to comprehensively study the long context understanding capabilities of these models in the tool calling setup. We modify existing benchmarks for challenge 1 and 3, and create a new evaluation set for challenge 2 to enable this analysis. We gradually increase the input context length and also vary the position of the answer in the input. When evaluated with several long context models, we observe a performance drop of 7% to 85% as the number of tools increases, a 7% to 91% degradation in answer retrieval as the tool responses length increases, and 13% and 40% degradation for as multi-turn conversations get longer. Our study shows that LLMs still struggle with long context in tool calling settings, motivating future research to drive further LLM improvements.

cs.SE↗

arXiv2024

Aligners: Decoupling LLMs and Alignment

Large Language Models (LLMs) need to be aligned with human expectations to ensure their safety and utility in most applications. Alignment is challenging, costly, and needs to be repeated for every LLM and alignment criterion. We propose to decouple LLMs and alignment by training aligner models that can be used to align any LLM for a given criteria on an as-needed basis, thus also reducing the potential negative impacts of alignment on performance. Our recipe for training the aligner models solely relies on synthetic data generated with a (prompted) LLM and can be easily adjusted for a variety of alignment criteria. We use the same synthetic data to train inspectors, binary miss-alignment classification models to guide a "squad" of multiple aligners. Our empirical results demonstrate consistent improvements when applying aligner squad to various LLMs, including chat-aligned models, across several instruction-following and red-teaming datasets.

cs.CL↗

arXiv2024

Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks

Large language models (LLMs) have recently shown tremendous promise in serving as the backbone to agentic systems, as demonstrated by their performance in multi-faceted, challenging benchmarks like SWE-Bench and Agent-Bench. However, to realize the true potential of LLMs as autonomous agents, they must learn to identify, call, and interact with external tools and application program interfaces (APIs) to complete complex tasks. These tasks together are termed function calling. Endowing LLMs with function calling abilities leads to a myriad of advantages, such as access to current and domain-specific information in databases and knowledge sources, and the ability to outsource tasks that can be reliably performed by tools, e.g., a Python interpreter or calculator. While there has been significant progress in function calling with LLMs, there is still a dearth of open models that perform on par with proprietary LLMs like GPT, Claude, and Gemini. Therefore, in this work, we introduce the GRANITE-20B-FUNCTIONCALLING model under an Apache 2.0 license. The model is trained using a multi-task training approach on seven fundamental tasks encompassed in function calling, those being Nested Function Calling, Function Chaining, Parallel Functions, Function Name Detection, Parameter-Value Pair Detection, Next-Best Function, and Response Generation. We present a comprehensive evaluation on multiple out-of-domain datasets comparing GRANITE-20B-FUNCTIONCALLING to more than 15 other best proprietary and open models. GRANITE-20B-FUNCTIONCALLING provides the best performance among all open models on the Berkeley Function Calling Leaderboard and fourth overall. As a result of the diverse tasks and datasets used for training our model, we show that GRANITE-20B-FUNCTIONCALLING has better generalizability on multiple tasks in seven different evaluation datasets.

cs.LG↗

arXiv2024

Granite Code Models: A Family of Open Foundation Models for Code Intelligence

Large Language Models (LLMs) trained on code are revolutionizing the software development process. Increasingly, code LLMs are being integrated into software development environments to improve the productivity of human programmers, and LLM-based agents are beginning to show promise for handling complex tasks autonomously. Realizing the full potential of code LLMs requires a wide range of capabilities, including code generation, fixing bugs, explaining and documenting code, maintaining repositories, and more. In this work, we introduce the Granite series of decoder-only code models for code generative tasks, trained with code written in 116 programming languages. The Granite Code models family consists of models ranging in size from 3 to 34 billion parameters, suitable for applications ranging from complex application modernization tasks to on-device memory-constrained use cases. Evaluation on a comprehensive set of tasks demonstrates that Granite Code models consistently reaches state-of-the-art performance among available open-source code LLMs. The Granite Code model family was optimized for enterprise software development workflows and performs well across a range of coding tasks (e.g. code generation, fixing and explanation), making it a versatile all around code model. We release all our Granite Code models under an Apache 2.0 license for both research and commercial use.

cs.AI↗

arXiv2024

Modeling surface vibrations and their role in molecular adsorption: a generalized Langevin approach

The atomic vibrations of a solid surface can play a significant role in the reactions of surface-bound molecules, as well as their adsorption and desorption. Relevant phonon modes can involve the collective motion of atoms over a wide array of length scales. In this manuscript, we demonstrate how the generalized Langevin equation can be utilized to describe these collective motions weighted by their coupling to individual sites. Our approach builds upon the generalized Langevin oscillator (GLO) model originally developed by Tully \textit{et al.} We extend the GLO by deriving parameters from atomistic simulation data. We apply this approach to study the memory kernel of a model platinum surface and demonstrate that the memory kernel has a bimodal form due to coupling to both low-energy acoustic modes and high-energy modes near the Debye frequency. The same bimodal form was observed across a wide variety of solids of different elemental compositions, surface structures, and solvation states. By studying how these dominant modes depend on simulation size, we argue that the acoustic modes are frozen in the limit of macroscopic lattices. By simulating periodically replicated slabs of various sizes we quantify the influence of phonon confinement effects in the memory kernel and their concomitant effect on simulated sticking coefficients.

physics.chem-ph↗

arXiv2024

Structured Code Representations Enable Data-Efficient Adaptation of Code Language Models

Current language models tailored for code tasks often adopt the pre-training-then-fine-tuning paradigm from natural language processing, modeling source code as plain text. This approach, however, overlooks the unambiguous structures inherent in programming languages. In this work, we explore data-efficient adaptation of pre-trained code models by further pre-training and fine-tuning them with program structures. Specifically, we represent programs as parse trees -- also known as concrete syntax trees (CSTs) -- and adapt pre-trained models on serialized CSTs. Although the models that we adapt have been pre-trained only on the surface form of programs, we find that a small amount of continual pre-training and fine-tuning on CSTs without changing the model architecture yields improvements over the baseline approach across various code tasks. The improvements are found to be particularly significant when there are limited training examples, demonstrating the effectiveness of integrating program structures with plain-text representation even when working with backbone models that have not been pre-trained with structures.

cs.CL↗

arXiv2023

Explain-then-Translate: An Analysis on Improving Program Translation with Self-generated Explanations

This work explores the use of self-generated natural language explanations as an intermediate step for code-to-code translation with language models. Across three types of explanations and 19 programming languages constructed from the MultiPL-E dataset, we find the explanations to be particularly effective in the zero-shot case, improving performance by 12% on average. Improvements with natural language explanations are particularly pronounced on difficult programs. We release our dataset, code, and canonical solutions in all 19 languages.

cs.CL↗

arXiv2023

An Investigation of Representation and Allocation Harms in Contrastive Learning

The effect of underrepresentation on the performance of minority groups is known to be a serious problem in supervised learning settings; however, it has been underexplored so far in the context of self-supervised learning (SSL). In this paper, we demonstrate that contrastive learning (CL), a popular variant of SSL, tends to collapse representations of minority groups with certain majority groups. We refer to this phenomenon as representation harm and demonstrate it on image and text datasets using the corresponding popular CL methods. Furthermore, our causal mediation analysis of allocation harm on a downstream classification task reveals that representation harm is partly responsible for it, thus emphasizing the importance of studying and mitigating representation harm. Finally, we provide a theoretical explanation for representation harm using a stochastic block model that leads to a representational neural collapse in a contrastive learning setting.

stat.ML↗

arXiv2023

Fairness Evaluation in Text Classification: Machine Learning Practitioner Perspectives of Individual and Group Fairness

Mitigating algorithmic bias is a critical task in the development and deployment of machine learning models. While several toolkits exist to aid machine learning practitioners in addressing fairness issues, little is known about the strategies practitioners employ to evaluate model fairness and what factors influence their assessment, particularly in the context of text classification. Two common approaches of evaluating the fairness of a model are group fairness and individual fairness. We run a study with Machine Learning practitioners (n=24) to understand the strategies used to evaluate models. Metrics presented to practitioners (group vs. individual fairness) impact which models they consider fair. Participants focused on risks associated with underpredicting/overpredicting and model sensitivity relative to identity token manipulations. We discover fairness assessment strategies involving personal experiences or how users form groups of identity tokens to test model fairness. We provide recommendations for interactive tools for evaluating fairness in text classification.

cs.HC↗

arXiv2023

Geometry-biased Transformers for Novel View Synthesis

We tackle the task of synthesizing novel views of an object given a few input images and associated camera viewpoints. Our work is inspired by recent 'geometry-free' approaches where multi-view images are encoded as a (global) set-latent representation, which is then used to predict the color for arbitrary query rays. While this representation yields (coarsely) accurate images corresponding to novel viewpoints, the lack of geometric reasoning limits the quality of these outputs. To overcome this limitation, we propose 'Geometry-biased Transformers' (GBTs) that incorporate geometric inductive biases in the set-latent representation-based inference to encourage multi-view geometric consistency. We induce the geometric bias by augmenting the dot-product attention mechanism to also incorporate 3D distances between rays associated with tokens as a learnable bias. We find that this, along with camera-aware embeddings as input, allows our models to generate significantly more accurate outputs. We validate our approach on the real-world CO3D dataset, where we train our system over 10 categories and evaluate its view-synthesis ability for novel objects as well as unseen categories. We empirically validate the benefits of the proposed geometric biases and show that our approach significantly improves over prior works.

cs.CV↗

arXiv2022

Better Together? An Evaluation of AI-Supported Code Translation

Generative machine learning models have recently been applied to source code, for use cases including translating code between programming languages, creating documentation from code, and auto-completing methods. Yet, state-of-the-art models often produce code that is erroneous or incomplete. In a controlled study with 32 software engineers, we examined whether such imperfect outputs are helpful in the context of Java-to-Python code translation. When aided by the outputs of a code translation model, participants produced code with fewer errors than when working alone. We also examined how the quality and quantity of AI translations affected the work process and quality of outcomes, and observed that providing multiple translations had a larger impact on the translation process than varying the quality of provided translations. Our results tell a complex, nuanced story about the benefits of generative code models and the challenges software engineers face when working with their outputs. Our work motivates the need for intelligent user interfaces that help software engineers effectively work with generative code models in order to understand and evaluate their outputs and achieve superior outcomes to working alone.

cs.HC↗

arXiv2022

Investigating Explainability of Generative AI for Code through Scenario-based Design

What does it mean for a generative AI model to be explainable? The emergent discipline of explainable AI (XAI) has made great strides in helping people understand discriminative models. Less attention has been paid to generative models that produce artifacts, rather than decisions, as output. Meanwhile, generative AI (GenAI) technologies are maturing and being applied to application domains such as software engineering. Using scenario-based design and question-driven XAI design approaches, we explore users' explainability needs for GenAI in three software engineering use cases: natural language to code, code translation, and code auto-completion. We conducted 9 workshops with 43 software engineers in which real examples from state-of-the-art generative AI models were used to elicit users' explainability needs. Drawing from prior work, we also propose 4 types of XAI features for GenAI for code and gathered additional design ideas from participants. Our work explores explainability needs for GenAI for code and demonstrates how human-centered approaches can drive the technical development of XAI in novel domains.

cs.HC↗

arXiv2021

COVID-19 India Dataset: Parsing COVID-19 Data in Daily Health Bulletins from States in India

While India has been one of the hotspots of COVID-19, data about the pandemic from the country has proved to be largely inaccessible at scale. Much of the data exists in unstructured form on the web, and limited aspects of such data are available through public APIs maintained manually through volunteer effort. This has proved to be difficult both in terms of ease of access to detailed data and with regards to the maintenance of manual data-keeping over time. This paper reports on our effort at automating the extraction of such data from public health bulletins with the help of a combination of classical PDF parsers and state-of-the-art machine learning techniques. In this paper, we will describe the automated data-extraction technique, the nature of the generated data, and exciting avenues of ongoing work.

cs.CL↗

arXiv2021

On sensitivity of meta-learning to support data

Meta-learning algorithms are widely used for few-shot learning. For example, image recognition systems that readily adapt to unseen classes after seeing only a few labeled examples. Despite their success, we show that modern meta-learning algorithms are extremely sensitive to the data used for adaptation, i.e. support data. In particular, we demonstrate the existence of (unaltered, in-distribution, natural) images that, when used for adaptation, yield accuracy as low as 4\% or as high as 95\% on standard few-shot image classification benchmarks. We explain our empirical findings in terms of class margins, which in turn suggests that robust and safe meta-learning requires larger margins than supervised learning.

cs.LG↗