Incremental Sequence Classification with Temporal Consistency

2025/05/22 by Lucas Maystre, Gabriel Barello, Maystre, Lucas +12
Computer Science · #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Data Classification #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2505.16548

openalex publication_date 2025/05/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We address the problem of incremental sequence classification, where predictions are updated as new elements in the sequence are revealed. Drawing on temporal-difference learning from reinforcement learning, we identify a temporal-consistency condition that successive predictions should satisfy. We leverage this condition to develop a novel loss function for training incremental sequence classifiers. Through a concrete example, we demonstrate that optimizing this loss can offer substantial gains in data efficiency. We apply our method to text classification tasks and show that it improves predictive accuracy over competing approaches on several benchmark datasets. We further evaluate our approach on the task of verifying large language model generations for correctness in grade-school math problems. Our results show that models trained with our method are better able to distinguish promising generations from unpromising ones after observing only a few tokens.

Citations

• Qwen2.5 Technical Report
• Fast Best-of-N Decoding via Speculative Rejection
• Deep End-to-End Survival Analysis with Temporal Consistency
• Better & Faster Large Language Models via Multi-token Prediction
• Looking Right is Sometimes Right: Investigating the Capabilities of Decoder-only LLMs for Sequence Labeling
• Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads
• Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations
• Controlled Decoding from Language Models
• Let's Verify Step by Step
• Vision-Language Models as Success Detectors
• On the Statistical Benefits of Temporal Difference Learning
• Solving math word problems with process- and outcome-based feedback
• Temporal Label Smoothing for Early Event Prediction
• Language Models (Mostly) Know What They Know
• MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge
• OPT: Open Pre-trained Transformer Language Models
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• Training Verifiers to Solve Math Word Problems
• HuggingFace's Transformers: State-of-the-art Natural Language Processing
• TEASER: Early and Accurate Time Series Classification
• Blockwise Parallel Decoding for Deep Autoregressive Models
• Generating Wikipedia by Summarizing Long Sequences
• Distributional Reinforcement Learning with Quantile Regression
• A Distributional Perspective on Reinforcement Learning
• Playing Atari with Deep Reinforcement Learning
• Parametric Return Density Estimation for Reinforcement Learning
• The meaning and use of the area under a receiver operating characteristic (ROC) curve.
• Token-Supervised Value Models for Enhancing Mathematical Problem-Solving Capabilities of Large Language Models
• Reinforcement Learning: An Introduction

Related

• Incremental Sequence Learning
• Incremental Sequence Labeling: A Tale of Two Shifts
• A Policy for Early Sequence Classification
• Semi-supervised Sequence Learning
• RepCL: Exploring Effective Representation for Continual Text Classification