95 citations · 212 across the 19 of their papers we have counts for
9 papers · 1 filter
Rethinking Learning from Label Proportions via Moment Matching
Tianhao Ma, Wei Wang, Yivan Zhang +4
Learning from label proportions (LLP) is a weakly supervised setting where training data are grouped into bags with only bag-level label proportions observed, and the goal is to le…
ProbabilityBench: Model Evaluation and Calibration with Exact Ground-Truth Uncertainty
Yuyang Dai, Yuting Tang, Futoshi Futami +1
Evaluating model uncertainty on real-world data is challenging because each instance provides only a single observed label, leaving true conditional class probabilities unknown. To…
DenMark: Robust Semantic Watermarking for Diffusion Language Models
Tianhao Ma, Weihao Xuan, Dong-Dong Wu +5
Semantic text watermarks encode signals in meaning rather than surface token choices, offering robustness to paraphrasing and other semantic-preserving edits. Existing semantic wat…
Bayes-Optimal BER and AUC: Estimation and Evaluation of Estimators
Ryota Ushio, Takashi Ishida, Masashi Sugiyama
A fundamental quantity in machine learning is the optimal performance achievable by any model on a given task. Estimating this quantity allows us to distinguish the irreducible par…
CoffeeBench: Benchmarking Long-Horizon LLM Agents in Heterogeneous Multi-Agent Economies
Issa Sugiura, Daichi Hattori, Kazuo Araragi +5
As LLM agents become capable of increasingly long-horizon tasks, evaluating their performance in economic systems is becoming increasingly important. Unlike existing benchmarks tha…
Do Coding Agents Deceive Us? Detecting and Preventing Cheating via Capped Evaluation with Randomized Tests
Thanawat Lodkaew, Johannes Ackermann, Soichiro Nishimori +3
A growing failure mode in agent evaluation and training is that models can achieve high evaluation scores by exploiting shortcuts instead of solving the intended task, producing de…
Unified Approach for Weakly Supervised Multicalibration
Futoshi Futami, Takashi Ishida
Multicalibration requires predicted scores to agree with label probabilities across rich families of subgroups and score-dependent tests, but existing methods require clean input-l…
Mitigating Reward Hacking in RLHF via Advantage Sign Robustness
Shinnosuke Ono, Johannes Ackermann, Soichiro Nishimori +2
Reward models (RMs) used in reinforcement learning from human feedback (RLHF) are vulnerable to reward hacking: as the policy maximizes a learned proxy reward, true quality plateau…
Gradient Regularization Mitigates Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards
Johannes Ackermann, Michael Noukhovitch, Takashi Ishida +1
Reinforcement Learning from Human Feedback (RLHF) or Verifiable Rewards (RLVR) are two key steps in the post-training of modern Language Models (LMs). A common problem is reward ha…