1 citations · 1 across the 13 of their papers we have counts for
13 papers
When Metropolis and Hastings Meet Bradley and Terry: Exact MCMC From Preference Voting
Ariel Smogorghevski, Nir Rosenfeld, Yaniv Romano
Sampling from distributions conditioned on desired semantic properties is an emerging challenge in modern generative modeling. Metropolis-Hastings (MH) provides a principled route…
Truncate Bad, Upweight Good: BoN-Style Distillation via Rank-Based Classification
Yarin Bar, Yaniv Romano
Inference-time selection methods, such as Best-of-N, improve generation by sampling a pool of candidates and selecting the top-ranked completion according to a reward model. Distil…
Semi-Supervised Hypothesis Testing by Betting on Predictions
Yaniv Tenzer, Elad Tolochinsky, Yaniv Romano
We introduce a testing-by-betting framework that leverages predictions on unlabeled data to enhance the power of sequential hypothesis testing. Given limited samples from the joint…
Valid Best-Model Identification for LLM Evaluation via Low-Rank Factorization
Elad Tolochinsky, Yaniv Tenzer, Yaniv Romano
Selecting the best large language model (LLM) for a fixed benchmark is often expensive, since exhaustive evaluation requires running every model on every example. Multi-armed bandi…
Selective Safety Steering via Value-Filtered Decoding
Bat-Sheva Einbinder, Hen Davidov, Yee Whye Teh +2
While large language models (LLMs) are trained to align with human values, their generations may still violate safety constraints. A growing line of work addresses this problem by…
Spend Where It Counts: Dynamic Budget Allocation for Multi-Turn LLM Evaluation
Shai Feldman, Yaniv Romano
Evaluating and predicting the performance of large language models (LLMs) in multi-turn conversational settings is critical yet computationally expensive; key events---e.g., jailbr…
Temporal Difference Calibration in Sequential Tasks: Application to Vision-Language-Action Models
Shelly Francis-Meretzki, Mirco Mutti, Yaniv Romano +1
Recent advances in vision-language-action (VLA) models for robotics have highlighted the importance of reliable uncertainty quantification in sequential tasks. However, assessing a…
Accelerating Speculative Decoding with Block Diffusion Draft Trees
Liran Ringel, Yaniv Romano
Speculative decoding accelerates autoregressive language models by using a lightweight drafter to propose multiple future tokens, which the target model then verifies in parallel.…
Testing For Distribution Shifts with Conditional Conformal Test Martingales
Shalev Shaer, Yarin Bar, Drew Prinster +1
We propose a sequential test for detecting arbitrary distribution shifts that allows conformal test martingales (CTMs) to work under a fixed, reference-conditional setting. Existin…
Prediction-Powered Semi-Supervised Learning with Online Power Tuning
Noa Shoham, Ron Dorfman, Shalev Shaer +2
Prediction-Powered Inference (PPI) is a recently proposed statistical inference technique for parameter estimation that leverages pseudo-labels on both labeled and unlabeled data t…
Learning a Continue-Thinking Token for Enhanced Test-Time Scaling
Liran Ringel, Elad Tolochinsky, Yaniv Romano
Test-time scaling has emerged as an effective approach for improving language model performance by utilizing additional compute at inference time. Recent studies have shown that ov…
Segment-Based Attention Masking for GPTs
Shahar Katz, Liran Ringel, Yaniv Romano +1
Modern Language Models (LMs) owe much of their success to masked causal attention, the backbone of Generative Pre-Trained Transformer (GPT) models. Although GPTs can process the en…
Semi-Supervised Risk Control via Prediction-Powered Inference
Bat-Sheva Einbinder, Liran Ringel, Yaniv Romano
The risk-controlling prediction sets (RCPS) framework is a general tool for transforming the output of any machine learning model to design a predictive rule with rigorous error ra…