Imagen de Portada

Choosy Babies Need One Coach: Inducing Mode-Seeking Behavior in BabyLlama with Reverse KL Divergence

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Shi, Shaozhen, Matusevych, Yevgen, Nissim, Malvina
Formato:	Preprint
Publicado:	2024
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

BabyLlama-2: Ensemble-Distilled Models Consistently Outperform Teachers With Limited Data
por: Tastet, Jean-Loup, et al.
Publicado: (2024)

Generating Completions for Broca's Aphasic Sentences Using Large Language Models
por: van Vaals, Sijbren, et al.
Publicado: (2024)

mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models
por: Lai, Huiyuan, et al.
Publicado: (2024)

IT5: Text-to-text Pretraining for Italian Language Understanding and Generation
por: Sarti, Gabriele, et al.
Publicado: (2022)

Puzzled By ChatGPT? No more! A Jigsaw Puzzle to Promote AI Literacy and Awareness
por: Padovani, Francesca, et al.
Publicado: (2026)

Is Child-Directed Language Optimized for Word Learning? A Computational Study of Verb Meaning Acquisition
por: Padovani, Francesca, et al.
Publicado: (2026)

Child-Directed Language Does Not Consistently Boost Syntax Learning in Language Models
por: Padovani, Francesca, et al.
Publicado: (2025)

TACLer: Tailored Curriculum Reinforcement Learning for Efficient Reasoning
por: Lai, Huiyuan, et al.
Publicado: (2026)

BabyLM Turns 3: Call for papers for the 2025 BabyLM workshop
por: Charpentier, Lucas, et al.
Publicado: (2025)

Visually Grounded Speech Models have a Mutual Exclusivity Bias
por: Nortje, Leanne, et al.
Publicado: (2024)

The mutual exclusivity bias of bilingual visually grounded speech models
por: Oneata, Dan, et al.
Publicado: (2025)

BabyLM Turns 4 and Goes Multilingual: Call for Papers for the 2026 BabyLM Workshop
por: Choshen, Leshem, et al.
Publicado: (2026)

Small Language Models Also Work With Small Vocabularies: Probing the Linguistic Abilities of Grapheme- and Phoneme-Based Baby Llamas
por: Bunzeck, Bastian, et al.
Publicado: (2024)

Multidimensional Consistency Improves Reasoning in Language Models
por: Lai, Huiyuan, et al.
Publicado: (2025)

When Harry Meets Superman: The Role of The Interlocutor in Persona-Based Dialogue Generation
por: Occhipinti, Daniela, et al.
Publicado: (2025)

Are BabyLMs Second Language Learners?
por: Edman, Lukas, et al.
Publicado: (2024)

A gentle push funziona benissimo: making instructed models in Italian via contrastive activation steering
por: Scalena, Daniel, et al.
Publicado: (2024)

Practising responsibility: Ethics in NLP as a hands-on course
por: Nissim, Malvina, et al.
Publicado: (2025)

BAMBI: Developing Baby Language Models for Italian
por: Suozzi, Alice, et al.
Publicado: (2025)

Can Model Uncertainty Function as a Proxy for Multiple-Choice Question Item Difficulty?
por: Zotos, Leonidas, et al.
Publicado: (2024)

Are You Doubtful? Oh, It Might Be Difficult Then! Exploring the Use of Model Uncertainty for Question Difficulty Estimation
por: Zotos, Leonidas, et al.
Publicado: (2024)

The Role of the Availability Heuristic in Multiple-Choice Answering Behaviour
por: Zotos, Leonidas, et al.
Publicado: (2026)

When Babies Teach Babies: Can student knowledge sharing outperform Teacher-Guided Distillation on small datasets?
por: Iyer, Srikrishna
Publicado: (2024)

Multi-property Steering of Large Language Models with Dynamic Activation Composition
por: Scalena, Daniel, et al.
Publicado: (2024)

BabyVision: Visual Reasoning Beyond Language
por: Chen, Liang, et al.
Publicado: (2026)

Child-directed speech facilitates production, not comprehension, in BabyLMs
por: Bunzeck, Bastian, et al.
Publicado: (2026)

Non Verbis, Sed Rebus: Large Language Models are Weak Solvers of Italian Rebuses
por: Sarti, Gabriele, et al.
Publicado: (2024)

BabyReasoningBench: Generating Developmentally-Inspired Reasoning Tasks for Evaluating Baby Language Models
por: Dhole, Kaustubh D.
Publicado: (2026)

CAIT: A Syntactic Parsing Toolkit for Child-Adult InTeractions
por: Padovani, Francesca, et al.
Publicado: (2026)

BabyHGRN: Exploring RNNs for Sample-Efficient Training of Language Models
por: Haller, Patrick, et al.
Publicado: (2024)

BAMBINO-LM: (Bilingual-)Human-Inspired Continual Pretraining of BabyLM
por: Shen, Zhewen, et al.
Publicado: (2024)

Unsupervised Word-level Quality Estimation for Machine Translation Through the Lens of Annotators (Dis)agreement
por: Sarti, Gabriele, et al.
Publicado: (2025)

Mini Minds: Exploring Bebeshka and Zlata Baby Models
por: Proskurina, Irina, et al.
Publicado: (2023)

BabyLM's First Constructions: Causal probing provides a signal of learning
por: Rozner, Joshua, et al.
Publicado: (2025)

CLASS-IT: Conversational and Lecture-Aligned Small-Scale Instruction Tuning for BabyLMs
por: Capone, Luca, et al.
Publicado: (2025)

BabyBabelLM: A Multilingual Benchmark of Developmentally Plausible Training Data
por: Jumelet, Jaap, et al.
Publicado: (2025)

Findings of the BabyLM Challenge: Sample-Efficient Pretraining on Developmentally Plausible Corpora
por: Warstadt, Alex, et al.
Publicado: (2025)

BabyLM's First Words: Word Segmentation as a Phonological Probing Task
por: Goriely, Zébulon, et al.
Publicado: (2025)

Do Construction Distributions Shape Formal Language Learning In German BabyLMs?
por: Bunzeck, Bastian, et al.
Publicado: (2025)

BabyLM Challenge: Exploring the Effect of Variation Sets on Language Model Training Efficiency
por: Haga, Akari, et al.
Publicado: (2024)