52 citations · 88 across the 22 of their papers we have counts for
28 papers · 1 filter
Reward-Gated On-Policy Distillation
Mohammad Sadegh Akhondzadeh, Vijay Lingam, Atula Tejaswi +3
On-policy distillation is a powerful way to transfer reasoning ability from a strong teacher to a smaller student: the student samples trajectories from its own policy, and the tea…
Test-Time Training Undermines Safety Guardrails
Simone Antonelli, Sadegh Akhondzadeh, Aleksandar Bojchevski
Test-Time Training (TTT) is an emerging paradigm that enables models to adapt their parameters during inference, improving performance on tasks such as few-shot learning, retrieval…
EvA: Evolutionary Attacks on Graphs
Mohammad Sadegh Akhondzadeh, Soroush H. Zargarbashi, Jimin Cao +1
Even a slight perturbation in the graph structure can cause a significant drop in the accuracy of graph neural networks (GNNs). Most existing attacks leverage gradient information…
One Sample is Enough to Make Conformal Prediction Robust
Soroush H. Zargarbashi, Mohammad Sadegh Akhondzadeh, Aleksandar Bojchevski
For any black-box model, conformal prediction (CP) returns prediction sets guaranteed to include the true label with high adjustable probability. Robust CP (RCP) extends the guaran…
Robust Conformal Prediction with a Single Binary Certificate
Soroush H. Zargarbashi, Aleksandar Bojchevski
Conformal prediction (CP) converts any model's output to prediction sets with a guarantee to cover the true label with (adjustable) high probability. Robust CP extends this guarant…
KurTail : Kurtosis-based LLM Quantization
Mohammad Sadegh Akhondzadeh, Aleksandar Bojchevski, Evangelos Eleftheriou +1
One of the challenges of quantizing a large language model (LLM) is the presence of outliers. Outliers often make uniform quantization schemes less effective, particularly in extre…