English
Related papers

Related papers: SPICE: Submodular Penalized Information-Conflict S…

200 papers

Uncertainty estimation is critical in high-stakes machine learning applications. One effective way to estimate uncertainty is conformal prediction, which can provide predictive inference with statistical coverage guarantees. We present a…

Machine Learning · Computer Science 2023-11-03 Nathaniel Diamant , Ehsan Hajiramezanali , Tommaso Biancalani , Gabriele Scalia

Large language models (LLMs) have shown strong reasoning capabilities when fine-tuned with reinforcement learning (RL). However, such methods require extensive data and compute, making them impractical under many realistic training budgets.…

Machine Learning · Computer Science 2026-04-17 Dai Do , Manh Nguyen , Svetha Venkatesh , Hung Le

In reinforcement learning for safety-critical settings, it is often desirable for the agent to obey safety constraints at all points in time, including during training. We present a novel neurosymbolic approach called SPICE to solve this…

Machine Learning · Computer Science 2023-03-01 Greg Anderson , Swarat Chaudhuri , Isil Dillig

We consider the problem of selecting an optimal subset of information sources for a hypothesis testing/classification task where the goal is to identify the true state of the world from a finite set of hypotheses, based on finite…

Machine Learning · Statistics 2024-07-01 Jayanth Bhargav , Mahsa Ghasemi , Shreyas Sundaram

Selecting appropriate training data is crucial for effective instruction fine-tuning of large language models (LLMs), which aims to (1) elicit strong capabilities, and (2) achieve balanced performance across a diverse range of tasks.…

Computation and Language · Computer Science 2025-01-22 Qirun Dai , Dylan Zhang , Jiaqi W. Ma , Hao Peng

Reward-model-based fine-tuning is a central paradigm in aligning Large Language Models with human preferences. However, such approaches critically rely on the assumption that proxy reward models accurately reflect intended supervision, a…

Computation and Language · Computer Science 2026-01-21 Zixuan Liu , Siavash H. Khajavi , Guangkai Jiang , Xinru Liu

Instruction tuning has unlocked powerful capabilities in large language models (LLMs), effectively using combined datasets to develop generalpurpose chatbots. However, real-world applications often require a specialized suite of skills…

Computation and Language · Computer Science 2024-06-14 Mengzhou Xia , Sadhika Malladi , Suchin Gururangan , Sanjeev Arora , Danqi Chen

Instruction tuning has underscored the significant potential of large language models (LLMs) in producing more human controllable and effective outputs in various domains. In this work, we focus on the data selection problem for…

Machine Learning · Computer Science 2025-09-01 Yang Wu , Huayi Zhang , Yizheng Jiao , Lin Ma , Xiaozhong Liu , Jinhong Yu , Dongyu Zhang , Dezhi Yu , Wei Xu

When different groups' values differ, one approach to model alignment is to steer models at inference time towards each group's preferences. However, techniques like in-context learning only consider similarity when drawing few-shot…

Computation and Language · Computer Science 2025-09-26 Quan Ze Chen , K. J. Kevin Feng , Chan Young Park , Amy X. Zhang

The similarity among samples and the discrepancy between clusters are two crucial aspects of image clustering. However, current deep clustering methods suffer from the inaccurate estimation of either feature similarity or semantic…

Computer Vision and Pattern Recognition · Computer Science 2022-11-23 Chuang Niu , Hongming Shan , Ge Wang

Although Large language Model (LLM)-powered information extraction (IE) systems have shown impressive capabilities, current fine-tuning paradigms face two major limitations: high training costs and difficulties in aligning with LLM…

Computation and Language · Computer Science 2025-12-16 Yushen Fang , Jianjun Li , Mingqian Ding , Chang Liu , Xinchi Zou , Wenqi Yang

Self-improving systems require environmental interaction for continuous adaptation. We introduce SPICE (Self-Play In Corpus Environments), a reinforcement learning framework where a single model acts in two roles: a Challenger that mines…

Computation and Language · Computer Science 2025-10-29 Bo Liu , Chuanyang Jin , Seungone Kim , Weizhe Yuan , Wenting Zhao , Ilia Kulikov , Xian Li , Sainbayar Sukhbaatar , Jack Lanchantin , Jason Weston

The task of infomin learning aims to learn a representation with high utility while being uninformative about a specified target, with the latter achieved by minimising the mutual information between the representation and the target. It…

Machine Learning · Computer Science 2023-02-22 Yanzhi Chen , Weihao Sun , Yingzhen Li , Adrian Weller

Models notoriously suffer from dataset biases which are detrimental to robustness and generalization. The identify-emphasize paradigm shows a promising effect in dealing with unknown biases. However, we find that it is still plagued by two…

Machine Learning · Computer Science 2022-11-29 Bowen Zhao , Chen Chen , Qian-Wei Wang , Anfeng He , Shu-Tao Xia

Direct preference optimization methods have emerged as a computationally efficient alternative to Reinforcement Learning from Human Feedback (RLHF) for aligning Large Language Models (LLMs). Latest approaches have streamlined the alignment…

Machine Learning · Computer Science 2026-02-04 Maksim Afanasyev , Illarion Iov

High-quality supervised fine-tuning (SFT) data are crucial for eliciting strong capabilities from pretrained large language models (LLMs). Typically, instructions are paired with multiple responses sampled from other LLMs, which are often…

Computation and Language · Computer Science 2026-01-13 Dylan Zhang , Qirun Dai , Hao Peng

We introduce and evaluate Stated Preference for Interaction and Continued Engagement (SPICE), a simple diagnostic signal elicited by asking a Large Language Model a YES or NO question about its willingness to re-engage with a user's…

Computation and Language · Computer Science 2025-09-23 Thomas Manuel Rost , Martina Figlia , Bernd Wallraff

The Bayesian and Akaike information criteria aim at finding a good balance between under- and over-fitting. They are extensively used every day by practitioners. Yet we contend they suffer from at least two afflictions: their penalty…

Statistics Theory · Mathematics 2026-03-20 Sylvain Sardy , Maxime van Cutsem , Sara van de Geer

Recent work shows that in-context learning and optimization of in-context examples (ICE) can significantly improve the accuracy of large language models (LLMs) on a wide range of tasks, leading to an apparent consensus that ICE optimization…

Computation and Language · Computer Science 2024-06-07 Pragya Srivastava , Satvik Golechha , Amit Deshpande , Amit Sharma

Deep ensembles perform better than a single network thanks to the diversity among their members. Recent approaches regularize predictions to increase diversity; however, they also drastically decrease individual members' performances. In…

Machine Learning · Computer Science 2021-01-15 Alexandre Rame , Matthieu Cord
‹ Prev 1 2 3 10 Next ›