← Search

Roland Memisevic

14 accepted papers

2026

Can Vision-Language Models Answer Face to Face Questions in the Real-World?

ICLR 2026poster

AI models have made significant strides in recent years in their ability to describe and answer questions about real-world images. They have also made progress in the ability to converse with users in real-time using audio input. This raises the question: have we reached the point where AI models, c…

Cited by 0SourceScholar
2026

Enhancing Hallucination Detection through Noise Injection

ICLR 2026poster

Large Language Models (LLMs) are prone to generating plausible yet incorrect responses, known as hallucinations. Effectively detecting hallucinations is therefore crucial for the safe deployment of LLMs. Recent research has linked hallucinations to model uncertainty, suggesting that hallucinations…

Cited by 0SourceScholar
2026

From Out-of-Distribution Detection to Hallucination Detection: A Geometric View

ICML 2026poster

Detecting hallucinations in large language models is a critical open problem with significant implications for safety and reliability. While existing hallucination detection methods achieve strong performance in question‑answering tasks, they remain less effective on tasks requiring reasoning. In th…

Cited by 0SourceScholar
2025

Can Multi-Modal LLMs Provide Live Step-by-Step Task Guidance?

NeurIPS 2025poster

Multi-modal Large Language Models (LLM) have advanced conversational abilities but struggle with providing live, interactive step-by-step guidance, a key capability for future AI assistants. Effective guidance requires not only delivering instructions but also detecting their successful execution, a…

Cited by 0SourcecodeScholar
2025

Multi-Draft Speculative Sampling: Canonical Decomposition and Theoretical Limits

ICLR 2025spotlight

We consider multi-draft speculative sampling, where the proposal sequences are sampled independently from different draft models. At each step, a token-level draft selection scheme takes a list of valid tokens as input and produces an output token whose distribution matches that of the target mode…

Cited by 0SourcePDFScholar
2024

ClevrSkills: Compositional Language And Visual Reasoning in Robotics

NeurIPS 2024poster

Robotics tasks are highly compositional by nature. For example, to perform a high-level task like cleaning the table a robot must employ low-level capabilities of moving the effectors to the objects on the table, pick them up and then move them off the table one-by-one, while re-evaluating the conse…

2024

Look, Remember and Reason: Grounded Reasoning in Videos with Language Models

ICLR 2024poster

Multi-modal language models (LM) have recently shown promising performance in high-level reasoning tasks on videos. However, existing methods still fall short in tasks like causal or compositional spatiotemporal reasoning over actions, in which model predictions need to be grounded in fine-grained…

Cited by 18SourcePDFScholar
2024

What to Say and When to Say it: Live Fitness Coaching as a Testbed for Situated Interaction

NeurIPS 2024poster

Vision-language models have shown impressive progress in recent years. However, existing models are largely limited to turn-based interactions, where each turn must be stepped (i.e., prompted) by the user. Open-ended, asynchronous interactions, where an AI model may proactively deliver timely respon…

2023

Deductive Verification of Chain-of-Thought Reasoning

NeurIPS 2023poster

Large Language Models (LLMs) significantly benefit from Chain-of-thought (CoT) prompting in performing various reasoning tasks. While CoT allows models to produce more comprehensive reasoning processes, its emphasis on intermediate reasoning steps can inadvertently introduce hallucinations and accum…

2017

The "Something Something" Video Database for Learning and Evaluating Visual Common Sense

ICCV 2017poster

Neural networks trained on datasets such as ImageNet have led to major advances in visual object classification. One obstacle that prevents networks from reasoning more deeply about complex scenes and situations, and from integrating visual knowledge with natural language, like humans do, is their l…

Cited by 1878PDFScholar
2016

Architectural Complexity Measures of Recurrent Neural Networks

NeurIPS 2016poster

In this paper, we systematically analyze the connecting architectures of recurrent neural networks (RNNs). Our main contribution is twofold: first, we present a rigorous graph-theoretic framework describing the connecting architectures of RNNs in general. Second, we propose three architecture comple…