8 papers
Effectiveness of Hidden-State Probes for Localizing Reasoning Failures
Shubham Subhnil, Yannis Belkhiter, Erik Miehling +2
Verifying the final answer of a long reasoning trace may show that the trace is wrong but it does not distinguish three cases. The problem may lie beyond the model's capability, an…
Local Causal Attribution of Chain-of-Thought Reasoning
Dennis Wei, Yannis Belkhiter, Erik Miehling +1
Understanding the causal structure of a language model's thought process is a problem of significant importance for both transparency and safety. In this work, we take a local appr…
TRACES: Tagging Reasoning Steps for Adaptive Cost-Efficient Early-Stopping
Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo +1
The field of Language Reasoning Models (LRMs) has been very active over the past few years with advances in training and inference techniques enabling LRMs to reason longer, and mo…
Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models
Yannis Belkhiter, Giulio Zizzo, Sergio Maffeis +2
The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by inv…
Step-Tagging: Toward controlling the generation of Language Reasoning Models through step monitoring
Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo +1
The field of Language Reasoning Models (LRMs) has been very active over the past few years with advances in training and inference techniques enabling LRMs to reason longer, and mo…
Dynamic Features Adaptation in Networking: Toward Flexible training and Explainable inference
Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo +2
As AI becomes a native component of 6G network control, AI models must adapt to continuously changing conditions, including the introduction of new features and measurements driven…
Pre-Hoc Predictions in AutoML: Leveraging LLMs to Enhance Model Selection and Benchmarking for Tabular datasets
Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo +2
The field of AutoML has made remarkable progress in post-hoc model selection, with libraries capable of automatically identifying the most performing models for a given dataset. Ne…
HarmLevelBench: Evaluating Harm-Level Compliance and the Impact of Quantization on Model Alignment
Yannis Belkhiter, Giulio Zizzo, Sergio Maffeis
With the introduction of the transformers architecture, LLMs have revolutionized the NLP field with ever more powerful models. Nevertheless, their development came up with several…