Image de couverture de livre

Beyond Tokens: Concept-Level Training Objectives for LLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Iyer, Laya, Somani, Pranav, Guo, Alice, Jurafsky, Dan, Shani, Chen
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models
par: Zhang, Christine, et autres
Publié: (2026)

From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning
par: Shani, Chen, et autres
Publié: (2025)

The Roots of Performance Disparity in Multilingual Language Models: Intrinsic Modeling Difficulty or Design Choices?
par: Shani, Chen, et autres
Publié: (2026)

Rethinking Word Similarity: Semantic Similarity through Classification Confusion
par: Zhou, Kaitlyn, et autres
Publié: (2025)

ELEPHANT: Measuring and understanding social sycophancy in LLMs
par: Cheng, Myra, et autres
Publié: (2025)

Cooking Up Creativity: Enhancing LLM Creativity through Structured Recombination
par: Mizrahi, Moran, et autres
Publié: (2025)

HumT DumT: Measuring and controlling human-like language in LLMs
par: Cheng, Myra, et autres
Publié: (2025)

Categorize Early, Integrate Late: Divergent Processing Strategies in Automatic Speech Recognition
par: Roll, Nathan, et autres
Publié: (2026)

Token-Level Uncertainty-Aware Objective for Language Model Post-Training
par: Liu, Tingkai, et autres
Publié: (2025)

Accommodation and Epistemic Vigilance: A Pragmatic Account of Why LLMs Fail to Challenge Harmful Beliefs
par: Cheng, Myra, et autres
Publié: (2026)

Trained on Tokens, Calibrated on Concepts: The Emergence of Semantic Calibration in LLMs
par: Nakkiran, Preetum, et autres
Publié: (2025)

A layer-wise analysis of Mandarin and English suprasegmentals in SSL speech models
par: de la Fuente, Antón, et autres
Publié: (2024)

Training LLMs Beyond Next Token Prediction -- Filling the Mutual Information Gap
par: Yang, Chun-Hao, et autres
Publié: (2025)

HEART: A Unified Benchmark for Assessing Humans and LLMs in Emotional Support Dialogue
par: Iyer, Laya, et autres
Publié: (2026)

SumTablets: A Transliteration Dataset of Sumerian Tablets
par: Simmons, Cole, et autres
Publié: (2026)

Othering and low status framing of immigrant cuisines in US restaurant reviews and large language models
par: Luo, Yiwei, et autres
Publié: (2023)

Humans overrely on overconfident language models, across languages
par: Rathi, Neil, et autres
Publié: (2025)

False Friends Are Not Foes: Investigating Vocabulary Overlap in Multilingual Language Models
par: Kallini, Julie, et autres
Publié: (2025)

Beyond Next Token Prediction: Patch-Level Training for Large Language Models
par: Shao, Chenze, et autres
Publié: (2024)

Distilling Token-Trained Models into Byte-Level Models
par: Bao, Zishuo, et autres
Publié: (2026)

Beyond Token-Level Policy Gradients for Complex Reasoning with Large Language Models
par: Xu, Mufan, et autres
Publié: (2026)

Transcribe, Translate, or Transliterate: An Investigation of Intermediate Representations in Spoken Language Models
par: Ògúnrèmí, Tolúlopé, et autres
Publié: (2025)

Data Checklist: On Unit-Testing Datasets with Usable Information
par: Zhang, Heidi C., et autres
Publié: (2024)

CausalGym: Benchmarking causal interpretability methods on linguistic tasks
par: Arora, Aryaman, et autres
Publié: (2024)

How Well Can LLMs Negotiate? NegotiationArena Platform and Analysis
par: Bianchi, Federico, et autres
Publié: (2024)

AnthroScore: A Computational Linguistic Measure of Anthropomorphism
par: Cheng, Myra, et autres
Publié: (2024)

Speculating LLMs' Chinese Training Data Pollution from Their Tokens
par: Zhang, Qingjie, et autres
Publié: (2025)

Dialect prejudice predicts AI decisions about people's character, employability, and criminality
par: Hofmann, Valentin, et autres
Publié: (2024)

Enhancing Character-Level Understanding in LLMs through Token Internal Structure Learning
par: Xu, Zhu, et autres
Publié: (2024)

Why 1 + 1 < 1 in Visual Token Pruning: Beyond Naive Integration via Multi-Objective Balanced Covering
par: Li, Yangfu, et autres
Publié: (2025)

Beyond Profile: From Surface-Level Facts to Deep Persona Simulation in LLMs
par: Wang, Zixiao, et autres
Publié: (2025)

Sentence-Level or Token-Level? A Comprehensive Study on Knowledge Distillation
par: Wei, Jingxuan, et autres
Publié: (2024)

A Benchmark for Learning to Translate a New Language from One Grammar Book
par: Tanzer, Garrett, et autres
Publié: (2023)

Vector Arithmetic in Concept and Token Subspaces
par: Feucht, Sheridan, et autres
Publié: (2025)

Beyond Single Concept Vector: Modeling Concept Subspace in LLMs with Gaussian Distribution
par: Zhao, Haiyan, et autres
Publié: (2024)

Verbalizing LLMs' assumptions to explain and control sycophancy
par: Cheng, Myra, et autres
Publié: (2026)

ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping
par: Chen, Shuang, et autres
Publié: (2025)

Informed Routing in LLMs: Smarter Token-Level Computation for Faster Inference
par: Han, Chao, et autres
Publié: (2025)

Token-Level Policy Optimization: Linking Group-Level Rewards to Token-Level Aggregation via Sequence-Level Likelihood
par: Lin, Xingyu, et autres
Publié: (2026)

Measuring Mental Health Variables in Computational Research: Toward Validated, Dimensional, and Transdiagnostic Approaches
par: Shani, Chen, et autres
Publié: (2025)