arXiv Machine Learning
Learning a Ranking from Human Feedback in Log-Concave Random Utility Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Statistics ML
Aug 25
Learning to Select and Rank from Choice-Based Feedback: A Simple Nested Approach
The paper tackles a ranking and selection problem where a company learns from choice-based feedback presented in dynamic assortments. It introduces two efficient algorithms—Nested Elimination for best-item identification and Nested Partition for full-ranking identification—each with instance-specific, non-asymptotic sample-complexity guarantees that are asymptotically worst-case optimal. The authors analyze the algorithms via multi-dimensional random walks, extend the framework to capacity-constrained displays, and validate their results with synthetic and real data experiments.
arXiv AI
Oct 1
On the Complexity of Preference-Based Bandits
The paper investigates preference-based bandits where a learner selects pairs of arms and receives binary preference feedback modeled by Bradley–Terry. It introduces the locally sensitive eluder dimension, a new complexity measure for logistic preference feedback, and proposes the GINOP algorithm that uses log-loss confidence sets to balance optimism and exploration. The authors prove a first-order regret bound showing that learning with preference feedback can be as statistically efficient as learning from direct rewards, and they validate their theory with empirical experiments.
arXiv Machine Learning
Oct 21
Ranking Bandits for Carousel Interfaces with Observable Browsing Depth
arXiv:2610.05220v2 Announce Type: replace Abstract: Carousel interfaces allow a recommender system to directly observe how far a user has browsed. This signal distinguishes displayed but unclicked it...
arXiv Machine Learning
Oct 22
Cascading Bandits with Noisy Observations: Algorithms and Regret Bounds
arXiv:2610. 29381v1 Announce Type: new Abstract: We propose a novel stochastic cascading bandit setup where probing an ordered shortlist of $K$ out of $L$ arms incurs position-dependent costs, and the learner observes only noisy feedback regarding each arm's true output.
arXiv AI
Jul 3
Distributionally Robust Listwise Preference Optimization
arXiv:2607. 01715v1 Announce Type: new Abstract: Existing robust preference optimization for language-model alignment mainly studies pairwise supervision and places robustness at the dataset, prompt, or preference-pair level.
arXiv AI
Sep 15
Confident Rankings with Fewer Items: Adaptive LLM Evaluation with Continuous Scores
arXiv:2601.13885v2 Announce Type: replace-cross Abstract: Computerized Adaptive Testing (CAT) has proven effective for efficient LLM evaluation on multiple-choice benchmarks, but modern LLM evaluatio...