20 results for “Continual learning”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
This paper proposes a continual RL framework based on Continual Backpropagation for autonomous racing, able to train a generalistic policy on a set of tracks and fine-tune it within 15 minutes to outp…
Haoyu Zhao, Sitian Tu, Yuxuan Huang, Huichi Zhou +15 more
This paper introduces Continual Learning Exam (CLE), a large-scale benchmark to measure experience-driven improvement in LLM agents.
The paper proposes CLASP, an end-to-end system with IMC acceleration for continual learning on edge platforms, addressing challenges of noisy computation and poor support for resource-efficient traini…
The paper introduces AGENTCL, a rigorous evaluation framework that uses controlled task streams to accurately measure an agent's ability to accumulate and reuse knowledge across multiple tasks, thereb…
This paper characterizes continual classification in homogeneous models as sequential projections and identifies regularity properties for local linear convergence.
Ran Liu, Min Yu, Mingqi Liu, Jianguo Jiang +6 more
The paper introduces AdvCL, a framework that repurposes adversarial perturbations as a geometric control signal to stabilize continual learning in large language models, significantly reducing forgett…
TailLoR is a new parameter-efficient finetuning method that uses the singular bases of pre-trained weights to learn low-rank updates, specifically penalizing updates along dominant directions to impro…
Purab Seth, Neil Shah, Kunal Jha, Samuel J. Gershman +2 more
The paper introduces Banyan, a new continual reinforcement learning benchmark, demonstrating that while task diversity enables local transfer across distribution shifts, it does not guarantee sustaine…
The paper proposes an active, continual learning paradigm for Vision-Language-Action models to improve fine-tuning efficiency.
The paper proposes Dynamic Adapter Routing (DAR), a novel method that significantly improves continual multimodal retrieval by adaptively selecting and merging specialized adapters.
This paper introduces MoDiCoL, a dataset for analyzing ASR robustness under real-world conditions and proposes a real-world-inspired continual learning curriculum.
Chaofan Pan, Lingfei Ren, Linbo Xiong, Yonghao Li +2 more
The paper proposes ReCAP, a novel continual learning framework for portfolio management, which adaptively combines policies from a library based on detected market regimes to achieve superior long-ter…
Cheng Gong, Haoyang Wang, Chao Lu, Zirui Li +1 more
This paper proposes Rollout-Retrieval Lifelong Policy Learning (R$^2$LPL), a framework for continual policy improvement in autonomous driving by retrieving corrective targets from recoverable mistakes…
This paper uses continual learning with explicit disfluency tokens to improve Automatic Speech Recognition (ASR) systems on disfluent speech, addressing the information loss and hallucinations caused…
Zahra Asadi, Haeseung Jeon, Sohyun Han, Md Mahmuduzzaman Kamol +2 more
FreeMOCA is a memory- and compute-efficient continual learning framework that uses adaptive layer-wise interpolation in parameter space to prevent catastrophic forgetting when analyzing evolving malwa…
This paper proposes a new method for continual learning that addresses catastrophic forgetting and explanation drift by anchoring regularization to class-consistent semantic concepts.
CLANE presents an end-to-end continual action recognition system deployed on neuromorphic hardware (Intel Loihi 2) using event cameras, achieving high accuracy with massive reductions in energy and la…
Yangxuan Zhou, Sha Zhao, Jiquan Wang, Shijian Li +1 more
EvoBrain proposes a dynamic, cross-task continual learning framework to overcome the limitations of task-specific EEG decoding, enabling unified and scalable brain-computer interfaces.
This paper studies how modular architecture and representational dimensionality affect compositional continual learning in a sequential A-B-A paradigm.
This paper investigates the geometric structure of recoverability in continual learning and challenges the Recoverability Diffusion hypothesis.