#incremental learning

try —

5 papers match

cs.SE2026

LimICE: Integrating LLM into ICE Framework for Efficient Loop Invariant Inference

Kai Fan, ShiWen Yu, GuangSheng Fan +3

The paper introduces LimICE, a tool that combines large language models with the ICE framework to incrementally generate sequences of lemmas as loop invariants, improving both succ…

#loop invariant synthesis#large language models#incremental learning#program verification
math.ST2026

Compactly supported radial basis functions as probability density functions

Sergio Díaz-Elbal, Andrei Martínez-Finkelshtein, Darío Ramos-López

The paper proposes using compactly supported radial basis functions, especially Wendland C² kernels, as a new parametric family of probability density functions and develops method…

#radial basis functions#probability density estimation#mixture models#incremental learning
cs.CV2026

Breaking the Model Forgetting Cycle in Long-Incremental 3D Object Detection

Peisheng Qian, Jie Xu, Xulei Yang +1

The paper proposes a Learning‑Dynamics‑driven Memory and Review (LDMR) framework to reduce forgetting in long‑incremental 3D object detection by monitoring per‑class detection qual…

#incremental learning#3d object detection#model forgetting#memory bank
cs.CV2026

Symbiosis-Inspired Knowledge Distillation for Incremental Object Detection

Mingyue Zeng, De Cheng, Zhipeng Xu +3

The paper introduces Symbiosis-Inspired Knowledge Distillation (SIKD), a method for incremental object detection that leverages spatial and semantic relationships between old and n…

#incremental learning#object detection#knowledge distillation#symbiosis
cs.AI2026

Mistake gating leads to energy and memory efficient continual learning

Aaron Pache, Mark CW van Rossum

The paper introduces memorized mistake‑gated learning, a biologically inspired rule that updates neural network weights only when classification errors occur, cutting the number of…

#continual learning#online learning#incremental learning#energy efficiency

One search, two signals: results blend meaning (embedding similarity, so papers that never use your words still surface) with keyword matches on titles, abstracts and summaries. Free, no sign-in needed.