Author name cluster

Rishabh Agrawal

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

7 papers

1 author row

AAMAS Conference 2026 Conference Paper

Agents of Diffusion: Enhancing Diffusion Language Models with Multi-Agent Reinforcement Learning for Structured Data Generation

Aja Khanal
Kaushik T. Ranade
Rishabh Agrawal
Kalyan S. Basu
Apurva Narayan

Generating high-quality structured data such as JSON records, remains a fundamental challenge for large language models (LLMs), particularlywhensemanticrichnessmustcoexistwithstrictschema adherence. While autoregressive LLMs offer strong structural consistency, they often struggle with semantic variation and output diversity. In contrast, diffusion language models (DLMs) introduce powerful mechanisms for semantic richness and bidirectional decoding, yet lack the inductive biases needed for reliable structure preservation. We present Agents of Diffusion (AoD), a novel framework that unifies the generative flexibility of DLMs with the reasoning capabilities of autoregressive models through languagemediated reinforcement learning. AoD frames structured text generation as a multi-agent alignment process, where a prompt optimization agent collaborates with a judge agent to iteratively guide a DLM using natural language feedback. This approach enables controllable, schema-consistent generation without modifying model parameters or relying on handcrafted constraints. AoD advances the state of controllable generation by demonstrating that diffusion models, when supervised by cooperative agents, can achieve both high semantic novelty and structural fidelity. Across multiple structured data benchmarks, AoD consistently outperforms diffusion and autoregressive baselines, establishing a new path forward for structure-aware, diversity-enhanced text synthesis. Code: https: //github. com/Idsl-group/AgentsOfDiffusion. Extended Version: https: //arxiv. org/abs/2601. 07152

PDF

AAMAS Conference 2026 Conference Paper

SCMRAG 2. 0: Efficient and Scalable Multi-hop Graph RAG with Multimodal Knowledge-Graphs and Agentic Self-Correction

Rishabh Agrawal
Uday Devulapalli
Apurva Narayan

We present SCMRAG 2. 0, a next-generation retrieval-augmented generation framework that unifies text, image, and structured data into a Multimodal Knowledge Graph. Unlike traditional graph RAG systems, SCMRAG 2. 0 introduces dual linkages via symbolic relations and cross-modal embeddings, an optimized graph-retrieval algorithm, and a multimodal agentic self-correction loop. By aligning language-level structure with vector-space signals and enabling agentic critique and repair, SCMRAG 2. 0 mitigates outdated context, incomplete reasoning chains, and hallucinations common in text-only graph RAG systems. Experiments on MMLU and MRAG- Bench demonstrate that SCMRAG 2. 0 significantly outperforms strong baselines like LightRAG in retrieval precision and factuality while maintaining computational efficiency.

PDF

AAAI Conference 2026 Short Paper

Towards Offline Imitation Learning: Strictly Batch Settings, Generalization, and Variability in Expertise

Rishabh Agrawal

My doctoral research develops a unified framework for offline imitation learning (IL) that tackles three central challenges: achieving sample efficiency in strictly batch settings, ensuring robustness and generalization under dynamics shifts, and learning from demonstrations of varying quality. At the core of this work is a new paradigm for strictly offline IL based on enforcing the Markov Balance Equation (MBE), a fundamental structural property of trajectory data. Using advanced conditional density estimation, I developed two algorithms, CKIL and MBIL, which achieve state-of-the-art performance in high-dimensional continuous-control tasks. Building upon this foundation, I developed the first Distributionally Robust Offline IL framework under a stationarity constraint, enabling robustness to transition-model mismatch without requiring any additional interaction. I am now extending this direction through Robust Behavior Foundation Models (RBFMs), which aim to generalize across dynamics shifts for a wide range of tasks. Finally, I propose a variational approach for learning from crowdsourced demonstrations by inferring and accounting for demonstrator expertise. Together, these contributions yield principled and practical IL algorithms with strong performance and robustness, broadening the applicability of IL to real-world domains such as robotics, healthcare, and autonomous systems.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Markov Balance Satisfaction Improves Performance in Strictly Batch Offline Imitation Learning

Rishabh Agrawal
Nathan Dahlin
Rahul Jain
Ashutosh Nayyar

Imitation learning (IL) is notably effective for robotic tasks where directly programming behaviors or defining optimal control costs is challenging. In this work, we address a scenario where the imitator relies solely on observed behavior and cannot make environmental interactions during learning. It does not have additional supplementary datasets beyond the expert's dataset nor any information about the transition dynamics. Unlike state-of-the-art (SOTA) IL methods, this approach tackles the limitations of conventional IL by operating in a more constrained and realistic setting. Our method uses the Markov balance equation and introduces a novel conditional density estimation-based imitation learning framework. It employs conditional normalizing flows for transition dynamics estimation and aims at satisfying a balance equation for the environment. Through a series of numerical experiments on Classic Control and MuJoCo environments, we demonstrate consistently superior empirical performance compared to many SOTA IL algorithms.

PDF Details DOI

AAMAS Conference 2025 Conference Paper

SCMRAG: Self-Corrective Multihop Retrieval Augmented Generation System for LLM Agents

Rishabh Agrawal
Murtaza Asrani
Hadi Youssef
Apurva Narayan

Existing Retrieval-Augmented Generation (RAG) systems primarily depend on static knowledge vectorstores which combine semantic similarity algorithms with reranking. This often leads to outdated information and retrieval errors. In this paper, we propose SCMRAG, a Self-Corrective Multihop Retrieval Augmented Generation system for LLM agents. We introduce an LLM-assisted dynamic knowledge graph creation step to enhance information retrieval and mitigate hallucinations. Unlike traditional RAG systems, SCMRAG includes a self-corrective agent driven mechanism that autonomously identifies and retrieves missing information from external web sources. Furthermore, SCMRAG’s internal reasoning agent determines whether the knowledge graph provides sufficient information or if a corrective step is needed. It further improves retrieval accuracy and efficiency. We benchmark the effectiveness of SCMRAG on five datasets - MultiHop-RAG, ARC AI2, PopQA, PubHealth, and WikiBio; showing significant improvements in retrieval precision and hallucination reduction across diverse tasks. Our results highlight SCMRAG’s potential to redefine how LLM agents interact with knowledge bases, offering a more adaptable and reliable solution for a wide range of applications.

PDF

NeurIPS Conference 2025 Conference Paper

SpecMAS: A Multi-Agent System for Self-Verifying System Generation via Formal Model Checking

Rishabh Agrawal
Kaushik Ranade
Aja Khanal
Kalyan Basu
Apurva Narayan

We present SpecMAS, a novel multi-agent system that autonomously constructs and formally verifies executable system models from natural language specifications. Given a Standard Operating Procedure (SOP) describing a target system, SpecMAS parses the specification, identifies relevant operational modes, variables, transitions, and properties, and generates a formal model in NuSMV code syntax, an industry-standard symbolic model checker. A dedicated reasoning agent extracts both explicit and implicit properties from the SOP, and verification is performed via temporal logic model checking. If any properties fail to verify, an autonomous debugging agent analyzes counterexamples and iteratively corrects the model until all properties are satisfied. This closed-loop system design guarantees provable correctness by construction and advances the state of the art in automated, interpretable, and deployable verification pipelines. We demonstrate the generality, correctness, and practical feasibility of SpecMAS across a set of representative case studies and propose a new benchmark dataset for the evaluation and comparison of model checking performance.

PDF Details

IJCAI Conference 2022 Conference Paper

Socially Intelligent Genetic Agents for the Emergence of Explicit Norms

Rishabh Agrawal
Nirav Ajmeri
Munindar Singh

Norms help regulate a society. Norms may be explicit (represented in structured form) or implicit. We address the emergence of explicit norms by developing agents who provide and reason about explanations for norm violations in deciding sanctions and identifying alternative norms. These agents use a genetic algorithm to produce norms and reinforcement learning to learn the values of these norms. We find that applying explanations leads to norms that provide better cohesion and goal satisfaction for the agents. Our results are stable for societies with differing attitudes of generosity.

PDF Details DOI