Machine Learning and Data Classification

Últimos artículos

• LESSER: Post-Training Data Selection with Output-Layer Gradients

  Lyuxin David Zhang, Eric Wong, Surbhi Goel, Anton Xue · 5 de octubre de 2026

  The choice of post-training data for large language models substantially affects downstream performance. Gradient-based data selection is a popular approach that ranks training data by how well their gradients align with those of a small validation set. However, ranking with full-parameter gradients…

• Planning to Learn

  Ian Osband · 5 de octubre de 2026

  Policy-gradient methods are central to modern reinforcement learning, including LLM post-training. When they struggle, the usual suspects are exploration, credit assignment and action-sampling noise. Classification has none of them. A classifier is a policy whose expected reward, its \emph{expected …

• Coverage You Can Steer: Online Conformal Calibration for RL-Driven Hardware-Aware NAS

  Pedro Brandimarte, Nerea Aranjuelo, Marcos Nieto, Oihana Otaegui · 5 de octubre de 2026

  Hardware-aware neural architecture search (NAS) is dominated by evaluation cost: every architecture must be trained before its reward is known. Conformal-prediction filters cut this cost by pruning candidates whose predicted-reward upper bound misses a threshold, with a distribution-free guarantee t…

• Probe the Harness: Setup Checks for Stale-Data RL Comparisons in Language Models

  Taiheng Pan · 5 de octubre de 2026

  Methods for training language models on stale samples are judged by comparisons against importance-corrected baselines. We show that details of the experimental harness can reverse the observed ranking of methods, and we introduce PTH (Probe The Harness), a set of checks that makes the harness visib…

• OptiSelect: How does the Optimizer Shape Data Curriculum?

  Simin Fan, Alireza Abdollahpoorrostam, Martin Jaggi · 5 de octubre de 2026

  Online data selection has demonstrated substantial efficiency gains for LLM pretraining by training on the most valuable candidates within each batch. Since a candidate's value is realized through its effective model update, principled selection should account for the optimizer step, which reshapes …

• Integrating Fairness and Explainability in a Multiple Instance Reinforcement Learning System

  Bente Hinkenhuis, Seyed Sahand Mohammadi Ziabari, Ali Mohammed Mansoor Alsahag · 2 de octubre de 2026

  Predicting student performance from educational interaction data requires models that are both accurate and sufficiently transparent to support meaningful intervention, while demographic information introduces an additional risk of unfair predictions. This study investigates a multi-objective framew…

• Effective Synthetic Data Curation Requires Group-Level Signals

  Cathy Jiao, Chenyan Xiong · 2 de octubre de 2026

  Synthetic data now is essential to LLM training, used to strengthen advanced capabilities such as autonomous and long-horizon task execution. Yet recent work shows that training on it at scale can degrade model generation, making it important to decide what synthetic data is worth training on. While…

• Scalable, Transferable Meta-network for Data Selection Requires a Different Loss (and Why the Obvious Choice is Problematic)

  Zilin Du, Bowen Yang, Boyang Albert Li · 2 de octubre de 2026

  Data selection is critical for training large language models on massive and heterogeneous corpora. Meta-learning for Training-data Selection offers a principled alternative to heuristic scoring by learning data weights from a target validation objective, but existing methods face a trade-off betwee…

• Parameter-Efficient Distributionally Robust Adaptation of Tabular Foundation Models under Subpopulation Shift

  Seonghwi Kim, Sung Ho Jo, Minwoo Chae · 2 de octubre de 2026

  Despite strong mean accuracy, tabular foundation models (TFMs) can perform poorly on underrepresented groups under subpopulation shift, where group proportions change between training and deployment. We propose DR-TFM, a parameter-efficient distributionally robust adaptation framework that requires …

• Optimal Transport Reweighting for Robust Learning under Spurious Correlations and Label Noise

  Sung Ho Jo, Seonghwi Kim, Wonsang Yun, Minwoo Chae · 2 de octubre de 2026

  Machine learning models often suffer performance degradation under subpopulation shift, particularly when spurious correlations cause models to rely on shortcut features that fail to generalize across subgroups. A recent line of work mitigates this issue by using loss-based signals to identify infor…

• Training-Aware Target Coverage for Synthetic Data Selection

  Yang Ba, Michelle V. Mancenido, Rong Pan · 2 de octubre de 2026

  Synthetic data are increasingly used to scale LLM training, yet more synthetic data do not necessarily produce better models. Useful synthetic data must add information relevant to the target task without introducing errors that offset their benefit, and the value of an example can change as the tra…

• ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization

  Sungho Park, Wonjoong Kim, Jue Zhang, Wook-Shin Han, Pengfei Gao, Chanyoung Park, Yongqiang Yao, Rao Fu, Elsie Nallipogu, Qingwei Lin, Victor R\"uhle · 2 de octubre de 2026

  Automated harness optimization can substantially improve LLM agents by iteratively updating their prompts, tool interfaces, and control logic from execution feedback. However, existing methods primarily optimize how the harness is updated while largely fixing which training scenarios generate the fe…

• A Generalisation Signal Need Not Be a Model-Selection Signal

  Aditya Nagarsekar, M P Ashish Bhat, Aadi Nesarkar, Vrishti Godhwani, Rahul Yedida, Aditya Challa, Danda Sravan, Snehanshu Saha · 1 de octubre de 2026

  Model selection in computational biology often relies on validation data drawn from the training regime, even when deployment lies outside it. When validation no longer preserves which model is best, a natural alternative is to rank candidates using properties of the trained network itself. We test …

• Signal-Routed Temperature Scaling: Low-Capacity Risk-Conditioned Calibration for Small Validation Budgets

  Wenhao Liang, Liangwei Nathan Zheng, Lin Yue, Wei Emma Zhang, Mingyu Guo, Olaf Maennel, Weitong Chen · 1 de octubre de 2026

  When a classifier is recalibrated from only a few thousand held-out examples, the capacity of the calibration map becomes a statistical design choice rather than a purely architectural one: a scalar map can underfit structured residual miscalibration, while a highly adaptive map can be hard to estim…

• LLMs as Adaptive Meta-Solvers: Strategy-Diverse RL for Industrial-Scale Optimization

  Shihao Zhang, Weiting Liu, Siyu Shao, Yitian Chen, Jianfeng Feng, Dongdong Ge, Yinyu Ye · 30 de septiembre de 2026

  Scaling LLM-based optimization from textbook-scale instances to real-world, industrial tasks remains a critical open challenge. Existing approaches are predominantly evaluated on small, self-contained textual problems and often commit to a solver-integrated paradigm, limiting their ability to handle…

• CPATTA: Conformal Supervision Allocation For Active Test-Time Adaptation

  Tingyu Shi, Fan Lyu, Haihua Zhu, Dadi Wang, Shaoliang Peng · 30 de septiembre de 2026

  Active Test-Time Adaptation (ATTA) improves model robustness under domain shift by selectively querying human annotations at deployment, but existing methods use heuristic uncertainty measures and suffer from low data selection efficiency, wasting human annotation budget. We propose Conformal Predic…

• From Checkpoint Variation to Selection Gains in Supervised Fine-Tuning

  Yupeng Chang, Wenxuan Zhang, Yuan Wu · 30 de septiembre de 2026

  Checkpoint selection is a routine decision in supervised fine-tuning (SFT): training produces multiple checkpoints, but only one is retained. Yet fixed-budget comparisons do not by themselves distinguish three empirical claims: whether more validation data improve checkpoint selection, whether a sel…

• When Noise Meets Long-Tail: Feature-Threshold Dual Calibration for Robust Pseudo-Labeling

  Ping Guo, Zhiqi Huang, Xinran Li · 29 de septiembre de 2026

  Pseudo-labeling has become a cornerstone of learning from unlabeled data in semantic segmentation. Yet its effectiveness drops sharply in real-world scenarios where strong imaging noise and long-tailed class distributions occur together. We trace this failure to a vicious cycle of pseudo-label degra…

• Refreshing Less, Selecting Better: Reusing Stale Gradient Features for Efficient Influence-Based Data Selection

  Jianchang Su, Yifan Zhang, Wei Zhang · 29 de septiembre de 2026

  Gradient-based data selection methods such as LESS score each candidate by the alignment between its gradient and a target validation gradient, and recomputing per-example gradient features at every new checkpoint dominates their cost. Across three selection seeds, two model families, two candidate …

• A Solvable Theory of Pre-training Data Poisoning: Regime-Dependent Scaling Exponents

  Indranil Halder, Rastri Dey, Cengiz Pehlevan · 29 de septiembre de 2026

  Pre-training data poisoning of large language models is usually studied using targeted backdoors and their survival through safety post-training, which leaves open a more basic question: how does a model's clean data performance degrade as the poison rate $\varepsilon$ grows? Motivated by our contro…

• GERIS: A Game-Theoretic Framework for Filtering Instance-Dependent Label Noise in License Plate Data Augmentation

  Seyedeh Sara Jalili Shani (Department of Computer Science, University of Alberta, Alberta, Canada), Rouhollah Ahmadian (Department of Mathematics and Computer Science, Amirkabir University of Technology, Iran), Amin Rahmani (Department of Mathematics and Computer Science, Amirkabir University of Technology, Iran), Mahdi Bideh (Department of Mathematics and Computer Science, Amirkabir University of Technology, Iran), Mehdi Ghatee (Department of Mathematics and Computer Science, Amirkabir University of Technology, Iran) · 29 de septiembre de 2026

  In this paper, we propose GERIS, a game-theoretic framework for instance selection in the data augmentation phase of license plate recognition systems. During augmentation, synthetic license plate images are generated and transformed using stochastic noise to simulate real-world conditions. However,…

• SpecRegMatch: Robust Semi-Supervised Regression for Vehicle Interior Noise Prediction

  Sejin Sim, Jinsoo Bae, Seoung Bum Kim · 29 de septiembre de 2026

  The rapid advancement of artificial intelligence has observed increased application in predicting vehicle interior noise levels within the automotive industry. However, the collection of labeled data for training models in this context involves significant costs. Previous studies in semi-supervised …

• Efficient Constrained Graph Search for Post-hoc Error Correction in Binary Classifiers

  Qinwu Xu · 28 de septiembre de 2026

  We introduce a model-agnostic framework for constrained post-hoc error correction in binary classifiers. Given a frozen base classifier, the method searches for an interpretable conjunction of feature--threshold rules that corrects residual false-positive or false-negative errors while explicitly co…

• Teacher-Anchored Selection of Post-Training Quantized Models under Domain Shift

  Alejandro Rodriguez Dominguez, Muhammad Shahzad, Xia Hong · 28 de septiembre de 2026

  Compressing a trained model yields a family of deployment candidates, and under domain shift the most compressed one need not be the one to deploy. We study selection over such a family, with candidates and teacher fixed and target labels absent or scarce. Two findings organize the label-free case. …

• Let Training Guide Selection: Online Synthetic Data Filtering via Real-Anchored Utility

  Yanran Wu, Sana Lakdawala, Renzo Tassara Miller, Chongyang Bai, Sharath Ciddu, Shivendra Pratap Singh, Kungang Li, Sandeep Pandey, Chunwei Liu · 25 de septiembre de 2026

  Synthetic data can scale training supervision when real-world data are limited, but noise and distribution mismatch can reduce its value. Existing synthetic data selection methods often emphasize fidelity or diversity rather than the learner's evolving needs. We propose FROST, an online framework th…