Rima Hazra

Do LLM Tutors Actually Teach? A Process-Centered Synthesis of the Tutoring Loop

Sayantan Adak, Rima Hazra, Koyel Mukherjee +2

Large language models can produce fluent tu- toring dialogue, but whether they actually help students learn remains unclear. We present a process-centered synthesis of LLM tutoring…

Structurally-bounded Agentic Graph Exploration for Evidence-Grounded Scholarly DeepSearch

Rima Hazra, Sayan Layek, Somnath Banerjee +2

We present Crase, a bounded and inspectable alternative to deep research agents for scholarly search. Instead of an open-ended search loop, Crase queries a search engine once for s…

Lost in Interpretation: The Plausibility-Faithfulness Trade-off in Cross-Lingual Explanations

Somnath Banerjee, Pranav Jha, Rima Hazra +1

LLMs deployed multilingually are often audited via English explanations for non-English inputs. We evaluate extractive explanations ''where the model identifies input token spans a…

SafeTutors: Benchmarking Pedagogical Safety in AI Tutoring Systems

Rima Hazra, Bikram Ghuku, Ilona Marchenko +5

Large language models are rapidly being deployed as AI tutors, yet current evaluation paradigms assess problem-solving accuracy and generic safety in isolation, failing to capture…

SafeMath: Safe Solutions for Unsafe Math Word Problems

Sagnik Basu, Subhrajit Mitra, Aman Juneja +3

Recent research points toward LLMs being manipulated through adversarial and seemingly benign inputs, resulting in harmful, biased, or policy-violating outputs. In this paper, we s…

Bridging the Multilingual Safety Divide: Efficient, Culturally-Aware Alignment for Global South Languages

Somnath Banerjee, Rima Hazra, Animesh Mukherjee

Large language models (LLMs) are being deployed across the Global South, where everyday use involves low-resource languages, code-mixing, and culturally specific norms. Yet safety…

cs.AI2026★ 2 cited

From Fluent to Verifiable: Claim-Level Auditability for Deep Research Agents

Razeen A Rasheed, Somnath Banerjee, Animesh Mukherjee +1

A deep research agent produces a fluent scientific report in minutes; a careful reader then tries to verify the main claims and discovers the real cost is not reading, but tracing:…

ProSocialAlign: Preference Conditioned Test Time Alignment in Language Models

Somnath Banerjee, Sayan Layek, Sayantan Adak +3

Current language model safety paradigms often fall short in emotionally charged or high-stakes settings, where refusal-only approaches may alienate users and naive compliance can a…

AURA: Affordance-Understanding and Risk-aware Alignment Technique for Large Language Models

Sayantan Adak, Pratyush Chatterjee, Somnath Banerjee +3

Present day LLMs face the challenge of managing affordance-based safety risks-situations where outputs inadvertently facilitate harmful actions due to overlooked logical implicatio…

Attributional Safety Failures in Large Language Models under Code-Mixed Perturbations

Somnath Banerjee, Pratyush Chatterjee, Shanu Kumar +4

While LLMs appear robustly safety-aligned in English, we uncover a catastrophic, overlooked weakness: attributional collapse under code-mixed perturbations. Our systematic evaluati…

cs.IR2025

MemeSense: An Adaptive In-Context Framework for Social Commonsense Driven Meme Moderation

Sayantan Adak, Somnath Banerjee, Rajarshi Mandal +4

Online memes are a powerful yet challenging medium for content moderation, often masking harmful intent behind humor, irony, or cultural symbolism. Conventional moderation systems…

Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment

Somnath Banerjee, Sayan Layek, Pratyush Chatterjee +2

Ensuring consistent safety across multiple languages remains a significant challenge for large language models (LLMs). We introduce Soteria, a lightweight yet powerful strategy tha…

Navigating the Cultural Kaleidoscope: A Hitchhiker's Guide to Sensitivity in Large Language Models

Somnath Banerjee, Sayan Layek, Hari Shrawgi +7

As LLMs are increasingly deployed in global applications, the importance of cultural sensitivity becomes paramount, ensuring that users from diverse backgrounds feel respected and…

SafeInfer: Context Adaptive Decoding Time Safety Alignment for Large Language Models

Somnath Banerjee, Sayan Layek, Soham Tripathy +3

Safety-aligned language models often exhibit fragile and imbalanced safety mechanisms, increasing the likelihood of generating unsafe content. In addition, incorporating new knowle…

Safety Arithmetic: A Framework for Test-time Safety Alignment of Language Models by Steering Parameters and Activations

Rima Hazra, Sayan Layek, Somnath Banerjee +1

Ensuring the safe alignment of large language models (LLMs) with human values is critical as they become integral to applications like translation and question answering. Current a…

Breaking Boundaries: Investigating the Effects of Model Editing on Cross-linguistic Performance

Somnath Banerjee, Avik Halder, Rajarshi Mandal +4

The integration of pretrained language models (PLMs) like BERT and GPT has revolutionized NLP, particularly for English, but it has also created linguistic imbalances. This paper s…

DistALANER: Distantly Supervised Active Learning Augmented Named Entity Recognition in the Open Source Software Ecosystem

Somnath Banerjee, Avik Dutta, Aaditya Agrawal +2

With the AI revolution in place, the trend for building automated systems to support professionals in different domains such as the open source software systems, healthcare systems…

cs.CL2024★ 3 cited

How (un)ethical are instruction-centric responses of LLMs? Unveiling the vulnerabilities of safety guardrails to harmful queries

Somnath Banerjee, Sayan Layek, Rima Hazra +1

In this study, we tackle a growing concern around the safety and ethical use of large language models (LLMs). Despite their potential, these models can be tricked into producing ha…