20 results for “self-supervised learning”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
This paper introduces a self-supervised framework for discovering novel, structured concepts from raw imagery without human annotation.
The paper proposes a time-aware self-supervised learning framework using BYOL to improve Android malware detection robustness by accurately accounting for app release times.
The authors improve the efficiency of reasoning models by fine-tuning them to predict their confidence in their answers, resulting in reduced generated tokens without sacrificing accuracy.
The paper introduces Noise-Aware BEATs (NABEATs), a noise-aware audio self-supervised learning framework that estimates clean BEATs representations from noisy audio signals using an auxiliary referenc…
Yifu Luo, Zeyu Chen, Haoyu Wang, Xinhao Hu +3 more
This paper introduces a new on-policy self-distillation framework tailored for diffusion large language models.
A novel self-supervised hard negative sampling technique is proposed for two-tower recommendation systems using a large language model to generate challenging and informative negatives.
The paper proposes pretraining a Perceiver-style in-context learner on synthetic data to solve Multiple Instance Learning (MIL) tasks efficiently in the low-label regime.
The paper introduces AutoHebb, a class of online learning feedforward neural networks that realize strong inductive biases and autonomously controlled learning.
Zhenting Qi, Susanna Maria Baby, Stefanie Anna Baby, Kan Yuan +4 more
The paper investigates the limits of self-evolution in LLM reasoning under closed-loop settings, finding that while self-improvement is significant, it consistently falls short of perfect oracle super…
Xiaobo Wang, Tong Wu, Min Tang, Jiaqi Li +2 more
The paper introduces SAVE, a framework that uses on-policy feedback and the value function to self-supervise and improve reward models, significantly enhancing RLHF performance across multiple benchma…
A new boosting algorithm that strong learns concept classes closed under O(log 1/γ)-XOR using O(log 1/ε) calls to a γ-advantage weak learner and additional samples, by connecting boosting with list-de…
Can Jin, Jiakang Li, Rui Wu, Eddy Zhang +1 more
The paper introduces Weak-Critic Strong Oversight, a method where a weak model guides a strong model's self-improvement by providing non-misleading revision directions, leading to scalable oversight.
This paper introduces survey sampling techniques to estimate or minimize empirical pairwise loss functions, showing that targeting informative pairs significantly reduces computational cost while main…
This paper extends gradient boosting to functions of vector inputs using a simple algorithm with histogram-based decision trees.
This paper investigates the use of strategically diverse data in self-training for LLMs, showing that diversity of approaches can matter more than correctness or teacher scale.
Weak self-training on synthetic data can amplify a language model's existing capabilities, but this effect is strictly dependent on the compatibility between the source and student models, not on the…
This paper presents Cohort Organized Learning (CoOL), a method for clustering data without explicit distance or similarity computations.
The paper identifies a linear predictive law linking the initial performance gap in on-policy self-distillation (OPSD) to the final performance improvement, allowing researchers to anticipate and tune…