paper-with-me

홈 › Papers

CurLL: A Developmental Framework to Evaluate Continual Learning in Language Models

2025-10-14 · Pavan Kalyan, Shubhra Mishra, Satya Lokam, Navin Goyal arxiv

We introduce a comprehensive continual learning dataset and benchmark (CurlL) grounded in human developmental trajectories from ages 5-10, enabling systematic and fine-grained assessment of models' ability to progressively acquire new skills. CurlL spans five developmental stages (0-4) covering ages 5-10, supported by a skill graph that breaks down broad skills into smaller abilities, concrete goals, and measurable indicators, while also capturing which abilities build on others. We generate a 23.4B-token synthetic dataset with controlled skill progression, vocabulary complexity, and format diversity, comprising paragraphs, comprehension-based QA (CQA), skill-testing QA (CSQA), and instruction-response (IR) pairs. Stage-wise token counts range from 2.12B to 6.78B tokens, supporting precise analysis of forgetting, forward transfer, and backward transfer. Using a 135M-parameter transformer trained under independent, joint, and sequential (continual) setups, we show trade-offs in skill retention and transfer efficiency. By mirroring human learning patterns and providing fine-grained control over skill dependencies, this work advances continual learning evaluations for language models.

📄 PDF Abstract BibTeX arXiv:2510.13008

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Similar Papers 제목 키워드 기반

Continual Developmental Neurosimulation Using Embodied Computational Agents

2021-03-07 · Bradly Alicea, Rishabh Chakrabarty, Stefan Dvoretskii, Akshara Gopi 외

There is much to learn through synthesis of Developmental Biology, Cognitive Science and Computational Modeling. Our path forward involves a design for developmentally-inspired learning agents based on Braitenberg Vehicl…

Active LearningContinual LearningDevelopmental Learning

A Retention-Centric Framework for Continual Learning with Guaranteed Model Developmental Safety

2024-10-04 · Gang Li, Wendi Yu, Yao Yao, Wei Tong 외

In real-world applications, learning-enabled systems often undergo iterative model development to address challenging or emerging tasks, which involve collecting new data, training a new model and validating the model. T…

Autonomous DrivingContinual Learningimage-classificationImage Classification+1

Multi-Scale Structural Features for Continual, Comprehensible Visual Recognition in a Developmental Learning Framework

2026-07-28 · Zeki Doruk Erden arxiv

Contemporary machine learning struggles to learn continually, reuse prior knowledge, and expose a comprehensible internal structure. A recently proposed developmental, gradient-free learning framework addresses these lim…

Growing Perspectives: Modelling Embodied Perspective Taking and Inner Narrative Development Using Large Language Models

2025-09-15 · Sabrina Patania, Luca Annese, Anna Lambiase, Anita Pellegrini 외 arxiv

Language and embodied perspective taking are essential for human collaboration, yet few computational models address both simultaneously. This work investigates the PerspAct system [1], which integrates the ReAct (Reason…

Developmental Predictive Coding Model for Early Infancy Mono and Bilingual Vocal Continual Learning

2024-12-23 · Xiaodan Chen, Alexandre Pitti, Mathias Quoy, Nancy F Chen

Understanding how infants perceive speech sounds and language structures is still an open problem. Previous research in artificial neural networks has mainly focused on large dataset-dependent generative models, aiming t…

Continual LearningLanguage Acquisition