Publications

Agent memoryACL 2026 (Main Conference, Long Papers), pages 16470–1648914 citations

APEX-MEM: Agentic Semi-Structured Memory with Temporal Reasoning for Long-Term Conversational AI

Pratyay Banerjee, Masud Moshtaghi, Shivashankar Subramanian, Amita Misra, Ankit Chadha

APEX-MEM uses an append-only property graph of temporally grounded, entity-centric events and a multi-tool retrieval agent to resolve evolving information at query time.

Research contributionIntroduced an append-only property graph memory that represents conversations as temporally grounded, entity-centric events and resolves evolving information at retrieval time.

Agent memoryEMNLP 2026, accepted, to appear

APEX-EM: Non-Parametric Online Learning for Autonomous Agents via Structured Procedural-Episodic Experience Replay

Pratyay Banerjee, Masud Moshtaghi, Ankit Chadha

APEX-EM introduced procedural knowledge graphs for LLM agent memory: its typed Procedural Knowledge Graph stores complete solution traces and retrieves them by meaning, operation structure, and graph traversal, with no model retraining.

Research contributionIntroduced procedural knowledge graphs for LLM agent memory through a typed Procedural Knowledge Graph, procedural-episodic experience replay, and cross-domain structural retrieval.

Multi-agentEMNLP 2026, accepted, to appear

Routed Graph Handoff: Adaptive Format Selection for Multi-Agent LLM Delegation

Pratyay Banerjee, Ankit Chadha

An extra classifier call picks per task whether one AI agent sends work to another as a dependency graph or as prose, gaining 12.7 points on τ-retail and preventing a 14.6-point loss on AppWorld.

Web-scale QAECML PKDD 2023 (Research Track), LNAI 14170, pages 593–61116 citations

Efficient Fine-Tuning Large Language Models for Knowledge-Aware Response Planning

Minh Nguyen, Kishan KC, Toan Nguyen, Ankit Chadha, Thuy Vu

The paper fine-tunes a question-answering model twice, once with retrieved web text and once without, so it can fall back on what it already knows; end-to-end accuracy rises 7.40% over a ranking baseline.

Web-scale QAFindings of ACL 2023, pages 14078–1409215 citations

Cross-Lingual Knowledge Distillation for Answer Sentence Selection in Low-Resource Languages

Shivanshu Gupta, Yoshitomo Matsubara, Ankit Chadha, Alessandro Moschitti

The paper trains answer-ranking models for languages with no labeled data by copying the soft scores of a strong English model, lifting Arabic accuracy on translated WikiQA from 64.2 to 76.3.

Controlled generationFindings of ACL 2023, pages 9440–945513 citations

Controlled Text Generation with Hidden Representation Transformations

Vaibhav Kumar, Hana Koorehdavoudi, Masud Moshtaghi, Amita Misra, Ankit Chadha, Emilio Ferrara

CHRT steers a language model away from toxic or negative text by editing its internal number vectors, cutting toxicity from 0.827 to 0.085 while adding only 0.01 seconds per generation.

Web-scale QAINTERSPEECH 2023, pages 3437–34413 citations

Question-Context Alignment and Answer-Context Dependencies for Effective Answer Sentence Selection

Minh Van Nguyen, Kishan KC, Toan Nguyen, Thien Huu Nguyen, Ankit Chadha, Thuy Vu

Instead of pasting neighboring sentences onto the question, this model matches question words to context words and builds a small sentence graph, raising top-1 answer accuracy on WikiQA from 68.09 to 74.16.

Federated learningNAACL 2022 (Main Conference), pages 2576–258627 citations

Training Mixed-Domain Translation Models via Federated Learning

Peyman Passban, Tanya Roosta, Rahul Gupta, Ankit Chadha, Clement Chung

Five domain-specific translation models use federated learning to build a mixed-domain neural machine translation system without pooling their data, with a method for selecting impactful parameters to control communication bandwidth.

Research contributionIntroduced a federated approach to mixed-domain neural machine translation and a bandwidth-control method that selects impactful parameters during federated updates.

Web-scale QACIKM 2022, pages 4707–471119 citations

WDRASS: A Web-scale Dataset for Document Retrieval and Answer Sentence Selection

Zeyu Zhang, Thuy Vu, Sunil Gandhi, Ankit Chadha, Alessandro Moschitti

WDRASS introduced a web-scale open-domain question answering dataset with 64,000 questions and 800,000+ labeled passages and sentences drawn from 30 million documents.

Research contributionIntroduced a web-scale dataset that connects document retrieval, passage reranking, and answer sentence selection using complete sentence-level answers rather than short answer matching.

Federated learningNeurIPS 2021 ENLSP Workshop11 citations

Communication-Efficient Federated Learning for Neural Machine Translation

Tanya Roosta, Peyman Passban, Ankit Chadha

To train translation models across separate data owners, the authors share only a few small added "Controller" layers, cutting exchanged parameters about six times with average translation quality slightly above the matched baseline.

SummarizationarXiv preprint Preprint

ACM -- Attribute Conditioning for Abstractive Multi Document Summarization

Aiswarya Sankar, Ankit Chadha

Adds a sentiment and polarity classifier to a news multi-document summarizer so it sticks to one viewpoint, raising the word-overlap ROUGE-1 score on MultiNews from 42.99 to 50.12 versus GraphSum.

Reading comprehensionarXiv preprint Preprint11 citations

BERTQA -- Attention on Steroids

Ankit Chadha, Rewa Sood

Adds two extra attention layers to BERT so a question and its passage look at each other directly, which lifts SQuAD 2.0 dev F1 from 74.15 to 77.03 on the base model.

SummarizationarXiv preprint Preprint

Deep Reinforced Self-Attention Masks for Abstractive Summarization (DR.SAS)

Ankit Chadha, Mohamed Masoud

Adds a reinforcement-learning agent that decides which input words a summarizer should ignore, raising the word-overlap score ROUGE-1 on CNN/Daily Mail news from 40.79 to 41.89 against fine-tuned UniLM.