Imagen de Portada

The Illusion of Latent Generalization: Bi-directionality and the Reversal Curse

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Coda-Forno, Julian, Wang, Jane X., Chaudhry, Arslan
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

CogBench: a large language model walks into a psychology lab
por: Coda-Forno, Julian, et al.
Publicado: (2024)

Reverse Training to Nurse the Reversal Curse
por: Golovneva, Olga, et al.
Publicado: (2024)

An Analysis and Mitigation of the Reversal Curse
por: Lv, Ang, et al.
Publicado: (2023)

Inducing anxiety in large language models can induce bias
por: Coda-Forno, Julian, et al.
Publicado: (2023)

The Factorization Curse: Which Tokens You Predict Underlie the Reversal Curse and More
por: Kitouni, Ouail, et al.
Publicado: (2024)

A Theoretical Analysis of Why Masked Diffusion Models Mitigate the Reversal Curse
por: Jeon, Moongyu, et al.
Publicado: (2026)

Mitigating Reversal Curse in Large Language Models via Semantic-aware Permutation Training
por: Guo, Qingyan, et al.
Publicado: (2024)

The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"
por: Berglund, Lukas, et al.
Publicado: (2023)

Exploring the Reversal Curse and Other Deductive Logical Reasoning in BERT and GPT-Based Large Language Models
por: Wu, Da, et al.
Publicado: (2023)

BiLD: Bi-directional Logits Difference Loss for Large Language Model Distillation
por: Li, Minchong, et al.
Publicado: (2024)

The Illusion of Insight in Reasoning Models
por: d'Aliberti, Liv G., et al.
Publicado: (2026)

The Consensus Trap: Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotation
por: Munir, Sheza, et al.
Publicado: (2026)

Speculative Decoding: Performance or Illusion?
por: Liu, Xiaoxuan, et al.
Publicado: (2025)

Recursive Concept Evolution for Compositional Reasoning in Large Language Models
por: Chaudhry, Sarim
Publicado: (2026)

Induction Head Toxicity Mechanistically Explains Repetition Curse in Large Language Models
por: Wang, Shuxun, et al.
Publicado: (2025)

Amphista: Bi-directional Multi-head Decoding for Accelerating LLM Inference
por: Li, Zeping, et al.
Publicado: (2024)

The Leaderboard Illusion
por: Singh, Shivalika, et al.
Publicado: (2025)

On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency
por: Wang, Yiming, et al.
Publicado: (2026)

Aligning to Illusions: Choice Blindness in Human and AI Feedback
por: Wu, Wenbin
Publicado: (2026)

An Illusion of Progress? Assessing the Current State of Web Agents
por: Xue, Tianci, et al.
Publicado: (2025)

The Illusion of Readiness in Health AI
por: Gu, Yu, et al.
Publicado: (2025)

Bi-directional Bias Attribution: Debiasing Large Language Models without Modifying Prompts
por: Lin, Yujie, et al.
Publicado: (2026)

Control Illusion: The Failure of Instruction Hierarchies in Large Language Models
por: Geng, Yilin, et al.
Publicado: (2025)

Reverse-Engineered Reasoning for Open-Ended Generation
por: Wang, Haozhe, et al.
Publicado: (2025)

Illusion or Algorithm? Investigating Memorization, Emergence, and Symbolic Processing in In-Context Learning
por: Niu, Jingcheng, et al.
Publicado: (2025)

Latent Abstraction for Retrieval-Augmented Generation
por: T, Ha Lan N., et al.
Publicado: (2026)

Do Depth-Grown Models Overcome the Curse of Depth? An In-Depth Analysis
por: Kapl, Ferdinand, et al.
Publicado: (2025)

Beyond the Needle's Illusion: Decoupled Evaluation of Evidence Access and Use under Semantic Interference at 326M-Token Scale
por: Lin, Tianwei, et al.
Publicado: (2026)

LEDOM: Reverse Language Model
por: Yin, Xunjian, et al.
Publicado: (2025)

Pareidolic Illusions of Meaning: ChatGPT, Pseudolaw and the Triumph of Form over Substance
por: McIntyre, Joe
Publicado: (2025)

Hallucination-Free Automatic Question & Answer Generation for Intuitive Learning
por: Wang, Nicholas X., et al.
Publicado: (2026)

The Illusion of Progress: Re-evaluating Hallucination Detection in LLMs
por: Janiak, Denis, et al.
Publicado: (2025)

The Curse of Recursion: Training on Generated Data Makes Models Forget
por: Shumailov, Ilia, et al.
Publicado: (2023)

Magellan: Guided MCTS for Latent Space Exploration and Novelty Generation
por: Chang, Lufan
Publicado: (2025)

Are UFOs Driving Innovation? The Illusion of Causality in Large Language Models
por: Carro, María Victoria, et al.
Publicado: (2024)

CorefInst: Leveraging LLMs for Multilingual Coreference Resolution
por: Arslan, Tuğba Pamay, et al.
Publicado: (2025)

A Rational Analysis of the Speech-to-Song Illusion
por: Marjieh, Raja, et al.
Publicado: (2024)

StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization
por: Wang, Shida, et al.
Publicado: (2023)

Generative Sentiment Analysis via Latent Category Distribution and Constrained Decoding
por: Zhou, Jun, et al.
Publicado: (2024)

Reversible Diffusion Decoding for Diffusion Language Models
por: Wang, Xinyun, et al.
Publicado: (2026)