11 papers
ToolHACE: A Dataset for Detecting Hallucinations in Tool-Based LLM Responses
Nikolay Ivanov, Sergey Pletenev, Elisei Rykov +8
Tool-augmented language models ground their responses in external tool outputs, improving reliability beyond parametric knowledge alone. However, even when the correct tool is call…
Synthetic Persona Pretraining: Alignment from Token Zero
Julian Minder, Viktor Moskvoretskii, Raghav Singhal +12
As language-model-based AI is increasingly deployed in autonomous settings, aligning its goals and values with those of humans becomes critical. Today, alignment, and the assistant…
Boosting Self-Consistency with Ranking
Maria Marina, Daniil Moskovskiy, Sergey Pletenev +3
Self-consistency improves large language models by sampling multiple reasoning paths and selecting the most frequent answer, but majority voting often fails to recover correct answ…
Tracing Persona Vectors Through LLM Pretraining
Viktor Moskvoretskii, Dominik Glandorf, Jorge Medina Moreira +2
How large language models internally represent high-level behaviors is a core interpretability question with direct relevance to AI safety: it determines what we can detect, audit,…
Evolutionary Search for Automated Design of Uncertainty Quantification Methods
Mikhail Seleznyov, Daniil Korbut, Viktor Moskvoretskii +3
Uncertainty quantification (UQ) methods for large language models are predominantly designed by hand based on domain knowledge and heuristics, limiting their scalability and genera…
Leveraging LLM Parametric Knowledge for Fact Checking without Retrieval
Artem Vazhentsev, Maria Marina, Daniil Moskovskiy +8
Trustworthiness is a core research challenge for agentic AI systems built on Large Language Models (LLMs). To enhance trust, natural language claims from diverse sources, including…
Will It Still Be True Tomorrow? Multilingual Evergreen Question Classification to Improve Trustworthy QA
Sergey Pletenev, Maria Marina, Nikolay Ivanov +6
Large Language Models (LLMs) often hallucinate in question answering (QA) tasks. A key yet underexplored factor contributing to this is the temporality of questions -- whether they…
LLM-Independent Adaptive RAG: Let the Question Speak for Itself
Maria Marina, Nikolay Ivanov, Sergey Pletenev +6
Large Language Models~(LLMs) are prone to hallucinations, and Retrieval-Augmented Generation (RAG) helps mitigate this, but at a high computational cost while risking misinformatio…
Do I look like a `cat.n.01` to you? A Taxonomy Image Generation Benchmark
Viktor Moskvoretskii, Alina Lobanova, Ekaterina Neminova +3
This paper explores the feasibility of using text-to-image models in a zero-shot setup to generate images for taxonomy concepts. While text-based methods for taxonomy enrichment ar…
Self-Taught Self-Correction for Small Language Models
Viktor Moskvoretskii, Chris Biemann, Irina Nikishina
Although large language models (LLMs) have achieved remarkable performance across various tasks, they remain prone to errors. A key challenge is enabling them to self-correct. Whil…
Challenges of Multi-Modal Coreset Selection for Depth Prediction
Viktor Moskvoretskii, Narek Alvandian
Coreset selection methods are effective in accelerating training and reducing memory requirements but remain largely unexplored in applied multimodal settings. We adapt a state-of-…