English
Related papers

Related papers: On a few pitfalls in KL divergence gradient estima…

200 papers

Many studies have evaluated the cognitive alignment of Pre-trained Language Models (PLMs), i.e., their correspondence to adult performance across a range of cognitive domains. Recently, the focus has expanded to the developmental alignment…

Computation and Language · Computer Science 2025-01-23 Raj Sanjay Shah , Sashank Varma

Seeking to improve model generalization, we consider a new approach based on distributionally robust learning (DRL) that applies stochastic gradient descent to the outer minimization problem. Our algorithm efficiently estimates the gradient…

Machine Learning · Statistics 2020-12-24 Soumyadip Ghosh , Mark Squillante

Calibration is commonly evaluated by comparing model confidence with its empirical correctness, implicitly treating reliability as a function of the confidence score alone. However, this view can hide substantial structure: models may be…

Machine Learning · Computer Science 2026-05-14 Katarzyna Kobalczyk , Mihaela van der Schaar

Deep Reinforcement Learning (DRL) has achieved remarkable success in domains requiring sequential decision-making, motivating its application to cybersecurity problems. However, transitioning DRL from laboratory simulations to bespoke cyber…

Large Language Models (LLMs) have shown remarkable progress, but their real-world application necessitates reliable calibration. This study conducts a comprehensive analysis of calibration degradation of LLMs across four dimensions: models,…

Computation and Language · Computer Science 2025-02-11 Hongseok Oh , Wonseok Hwang

Feedback alignment algorithms are an alternative to backpropagation to train neural networks, whereby some of the partial derivatives that are required to compute the gradient are replaced by random terms. This essentially transforms the…

Machine Learning · Computer Science 2023-06-06 Dominique Chu , Florian Bacho

The Kullback-Leibler (KL) divergence plays a central role in probabilistic machine learning, where it commonly serves as the canonical loss function. Optimization in such settings is often performed over the probability simplex, where the…

Machine Learning · Computer Science 2025-07-31 Adwait Datar , Nihat Ay

Existing agents for solving tasks such as ML engineering rely on prompting powerful language models. As a result, these agents do not improve with more experience. In this paper, we show that agents backed by weaker models that improve via…

Machine Learning · Computer Science 2025-09-04 Sherry Yang , Joy He-Yueya , Percy Liang

While imitation learning requires access to high-quality data, offline reinforcement learning (RL) should, in principle, perform similarly or better with substantially lower data quality by using a value function. However, current results…

Machine Learning · Computer Science 2024-10-30 Seohong Park , Kevin Frans , Sergey Levine , Aviral Kumar

Reparameterization (RP) and likelihood ratio (LR) gradient estimators are used throughout machine and reinforcement learning; however, they are usually explained as simple mathematical tricks without providing any insight into their nature.…

Machine Learning · Computer Science 2019-10-16 Paavo Parmas , Masashi Sugiyama

We study the depth of grade-school math (GSM) problem-solving capabilities of LLMs. To this end, we evaluate their performance on pairs of existing math word problems together so that the answer to the second problem depends on correctly…

Machine Learning · Computer Science 2024-10-03 Arian Hosseini , Alessandro Sordoni , Daniel Toyama , Aaron Courville , Rishabh Agarwal

Modern LLM RL systems separate rollout generation from policy optimization. These two stages are expected to produce token probabilities that match exactly. However, implementation differences can make them assign different values to the…

Machine Learning · Computer Science 2026-05-15 Tianle Zhong , Neiwen Ling , Yifan Pi , Zijun Wei , Tianshu Yu , Geoffrey Fox , Peng Wu , Xiao Yu

Large language models (LLMs) demonstrate remarkable performance on math word problems, yet they have been shown to struggle with meta-reasoning tasks such as identifying errors in student solutions. In this work, we investigate the…

Computation and Language · Computer Science 2025-09-03 KV Aditya Srivatsa , Kaushal Kumar Maurya , Ekaterina Kochmar

Large language models~(LLMs) have greatly advanced the frontiers of artificial intelligence, attaining remarkable improvement in model capacity. To assess the model performance, a typical approach is to construct evaluation benchmarks for…

Computation and Language · Computer Science 2023-11-06 Kun Zhou , Yutao Zhu , Zhipeng Chen , Wentong Chen , Wayne Xin Zhao , Xu Chen , Yankai Lin , Ji-Rong Wen , Jiawei Han

Gradient-based methods for instance-based explanation for large language models (LLMs) are hindered by the immense dimensionality of model gradients. In practice, influence estimation is restricted to a subset of model parameters to make…

Computation and Language · Computer Science 2026-01-30 Lukas Hinterleitner , Loris Schoenegger , Benjamin Roth

While deep learning and deep reinforcement learning (RL) systems have demonstrated impressive results in domains such as image classification, game playing, and robotic control, data efficiency remains a major challenge. Multi-task learning…

Machine Learning · Computer Science 2020-12-23 Tianhe Yu , Saurabh Kumar , Abhishek Gupta , Sergey Levine , Karol Hausman , Chelsea Finn

Reinforcement learning (RL) combines a control problem with statistical estimation: The system dynamics are not known to the agent, but can be learned through experience. A recent line of research casts `RL as inference' and suggests a…

Machine Learning · Computer Science 2020-11-05 Brendan O'Donoghue , Ian Osband , Catalin Ionescu

Reinforcement Learning (RL) for training Large Language Models is notoriously unstable. While recent studies attribute this to "training inference mismatch stemming" from inconsistent hybrid engines, standard remedies, such as Importance…

Machine Learning · Computer Science 2026-02-03 Yaxiang Zhang , Yingru Li , Jiacai Liu , Jiawei Xu , Ziniu Li , Qian Liu , Haoyuan Li

Large-scale distributed systems such as sensor networks, often need to achieve filtering and consensus on an estimated parameter from high-dimensional measurements. Running a Kalman filter on every node in such a network is computationally…

Optimization and Control · Mathematics 2017-04-12 Mathias Hudoba de Badyn , Mehran Mesbahi

Large language models (LLMs) could be valuable personal AI agents across various domains, provided they can precisely follow user instructions. However, recent studies have shown significant limitations in LLMs' instruction-following…

Artificial Intelligence · Computer Science 2025-03-31 Juyeon Heo , Miao Xiong , Christina Heinze-Deml , Jaya Narain