2 citations · 6 across the 33 of their papers we have counts for
38 papers
Beyond Shallow Alignment: How Post-Training Methods Determine Refusal Circuits And Steering Robustness
Hoang Cuong Nguyen, Mark Dras, Usman Naseem
How do the methods used to train language models to refuse harmful requests shape how that refusal actually works inside the model? We compare three post-training methods - supervi…
Pak3H: Evaluating the Cost of Cultural Mismatch in LLM Alignment with a Human-Contextualized Urdu Benchmark
Abdullah Hashmat, Usman Naseem, Agha Ali Raza
Large language models (LLMs) demonstrate strong Helpfulness, Harmlessness, and Honesty (3H) alignment in English-centric settings, but these gains transfer poorly to low-resource l…
SB-BEVFusion: Enhancing the Robustness against Sensor Malfunction and Corruptions
Markus Essl, Marta Moscati, Mubashir Noman +4
Multimodal sensor fusion has demonstrated remarkable performance improvements over unimodal approaches in 3D object detection for autonomous vehicles. Typically, existing methods t…
Towards Visually Grounded Multimodal Summarization via Cross-Modal Transformer and Gated Attention
Abid Ali, Diego Molla-Aliod, Usman Naseem
Multimodal summarization requires models to jointly understand textual and visual inputs to generate concise, semantically coherent summaries. Existing methods often inject shallow…
Measuring What Matters Beyond Text: Evaluating Multimodal Summaries by Quality, Alignment, and Diversity
Abid Ali, Diego Molla-Aliod, Usman Naseem
Multimodal Large Language Models (MLLMs) have facilitated Multimodal Summarization with Multimodal Output (MSMO), wherein systems generate concise textual summaries accompanied by…
Uno-Orchestra: Parsimonious Agent Routing via Selective Delegation
Zhiqing Cui, Haotong Xie, Jiahao Yuan +11
Large language model (LLM) multi-agent systems typically rely on rigid orchestration, committing either to flat per-query routing or to hand-engineered task decomposition, so decom…
Are Large Language Models Economically Viable for Industry Deployment?
Abdullah Mohammad, Sushant Kumar Ray, Pushkar Arora +5
Generative AI-powered by Large Language Models (LLMs)-is increasingly deployed in industry across healthcare decision support, financial analytics, enterprise retrieval, and conver…
AlignCultura: Towards Culturally Aligned Large Language Models?
Gautam Siddharth Kashyap, Mark Dras, Usman Naseem
Cultural alignment in Large Language Models (LLMs) is essential for producing contextually aware, respectful, and trustworthy outputs. Without it, models risk generating stereotype…
Over-Refusal and Representation Subspaces: A Mechanistic Analysis of Task-Conditioned Refusal in Aligned LLMs
Utsav Maskey, Mark Dras, Usman Naseem
Aligned language models that are trained to refuse harmful requests also exhibit over-refusal: they decline safe instructions that seemingly resemble harmful instructions. A natura…
Are Aligned Large Language Models Still Misaligned?
Usman Naseem, Gautam Siddharth Kashyap, Rafiq Ali +4
Misalignment in Large Language Models (LLMs) arises when model behavior diverges from human expectations and fails to simultaneously satisfy safety, value, and cultural dimensions,…
Can Large Language Models Make Everyone Happy?
Usman Naseem, Gautam Siddharth Kashyap, Ebad Shabbir +3
Misalignment in Large Language Models (LLMs) refers to the failure to simultaneously satisfy safety, value, and cultural dimensions, leading to behaviors that diverge from human ex…
When the Model Said 'No Comment', We Knew Helpfulness Was Dead, Honesty Was Alive, and Safety Was Terrified
Gautam Siddharth Kashyap, Mark Dras, Usman Naseem
Large Language Models (LLMs) need to be in accordance with human values-being helpful, harmless, and honest (HHH)-is important for safe deployment. Existing works use Supervised Fi…
Do Large Language Models Reflect Demographic Pluralism in Safety?
Usman Naseem, Gautam Siddharth Kashyap, Sushant Kumar Ray +3
Large Language Model (LLM) safety is inherently pluralistic, reflecting variations in moral norms, cultural expectations, and demographic contexts. Yet, existing alignment datasets…
Revealing the Truth with ConLLM for Detecting Multi-Modal Deepfakes
Gautam Siddharth Kashyap, Harsh Joshi, Niharika Jain +4
The rapid rise of deepfake technology poses a severe threat to social and political stability by enabling hyper-realistic synthetic media capable of manipulating public perception.…
Mechanistic Interpretability for Large Language Model Alignment: Progress, Challenges, and Future Directions
Usman Naseem
Large language models (LLMs) have achieved remarkable capabilities across diverse tasks, yet their internal decision-making processes remain largely opaque. Mechanistic interpretab…
Do Clinical Question Answering Systems Really Need Specialised Medical Fine Tuning?
Sushant Kumar Ray, Gautam Siddharth Kashyap, Sahil Tripathi +5
Clinical Question-Answering (CQA) industry systems are increasingly rely on Large Language Models (LLMs), yet their deployment is often guided by the assumption that domain-specifi…
CogMem: A Cognitive Memory Architecture for Sustained Multi-Turn Reasoning in Large Language Models
Yiran Zhang, Jincheng Hu, Mark Dras +1
Large language models (LLMs) excel at single-turn reasoning but often lose accuracy and coherence over extended, multi-turn interactions. Recent evaluations such as TurnBench highl…
Beyond the Black Box: Demystifying Multi-Turn LLM Reasoning with VISTA
Yiran Zhang, Mingyang Lin, Mark Dras +1
Recent research has increasingly focused on the reasoning capabilities of Large Language Models (LLMs) in multi-turn interactions, as these scenarios more closely mirror real-world…
Agentic Moderation: Multi-Agent Design for Safer Vision-Language Models
Juan Ren, Mark Dras, Usman Naseem
Agentic methods have emerged as a powerful and autonomous paradigm that enhances reasoning, collaboration, and adaptive control, enabling systems to coordinate and independently so…
SHIELD: Classifier-Guided Prompting for Robust and Safer LVLMs
Juan Ren, Mark Dras, Usman Naseem
Large Vision-Language Models (LVLMs) unlock powerful multimodal reasoning but also expand the attack surface, particularly through adversarial inputs that conceal harmful goals in…
Steering Over-refusals Towards Safety in Retrieval Augmented Generation
Utsav Maskey, Mark Dras, Usman Naseem
Safety alignment in large language models (LLMs) induces over-refusals -- where LLMs decline benign requests due to aggressive safety filters. We analyze this phenomenon in retriev…
Fairness Evaluation and Inference Level Mitigation in LLMs
Afrozah Nadeem, Mark Dras, Usman Naseem
Large language models often display undesirable behaviors embedded in their internal representations, undermining fairness, inconsistency drift, amplification of harmful content, a…
DUAL-Bench: Measuring Over-Refusal and Robustness in Vision-Language Models
Kaixuan Ren, Preslav Nakov, Usman Naseem
As vision-language models (VLMs) become increasingly capable, maintaining a balance between safety and usefulness remains a central challenge. Safety mechanisms, while essential, c…
Steering Towards Fairness: Mitigating Political Bias in LLMs
Afrozah Nadeem, Mark Dras, Usman Naseem
Recent advancements in large language models (LLMs) have enabled their widespread use across diverse real-world applications. However, concerns remain about their tendency to encod…