Formalizes Reasoning Portability (RP) and proposes RDB-CL to modulate per-sample KL regularization in RLVR for MLLM continual learning, achieving +12.0% Last accuracy over vanilla RLVR baseline by preserving reusable reasoning on high-RP samples.
A continual learning survey: Defying forgetting in classification tasks,
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
Fuzzy ARTMAP models are highly vulnerable to a new white-box attack aligned with their category competition, but progressive selective training yields stronger replay-free robustness than offline adversarial training under adaptive evaluation.
CapTrack shows post-training causes drift beyond facts, with instruction fine-tuning producing stronger behavioral changes than preference optimization across model families.
DeCoFlow decomposes normalizing flow subnets into frozen bases and low-rank adapters with alignment, auxiliary layers, and tail-aware loss to achieve continual anomaly detection with zero forgetting and few added parameters.
Diversity-aware memory policies improve test-time adaptation performance most under constrained memory budgets and challenging non-i.i.d. streams.
The relative rankings of continual learning methods are not preserved across different fine-tuning regimes defined by trainable parameter depth.
Proposes a self-evolving cognitive framework integrating causal world modeling, intervention-driven reasoning, and continual refinement for embodied scientific intelligence.
SF-NorMuon is a new schedule-free spectral optimizer that closes the gap with tuned AdamW on 125M-772M parameter models across 1-8x Chinchilla horizons while providing stationarity guarantees.
Task connectivity in graph-structured multi-task environments enhances generalization and stability, with stronger benefits for attention models than MLPs.
This is the first comprehensive survey of OOD generalization methodologies for time series, organized across data distribution, representation learning, and OOD evaluation.
Spatial Learning Entropy Maps derived from MLP weight adaptations during spatial pixel prediction tasks highlight image points with high learning impact.
DGMM is proposed as an explicit graph-structured memory architecture for AI that enables persistent episodic memory, cue-based recall, and context-dependent interpretation without retraining.
citing papers explorer
-
Reasoning Portability: Guiding Continual Learning for MLLMs in the RLVR Era
Formalizes Reasoning Portability (RP) and proposes RDB-CL to modulate per-sample KL regularization in RLVR for MLLM continual learning, achieving +12.0% Last accuracy over vanilla RLVR baseline by preserving reusable reasoning on high-RP samples.
-
CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training
CapTrack shows post-training causes drift beyond facts, with instruction fine-tuning producing stronger behavioral changes than preference optimization across model families.
-
DeCoFlow: Structural Decomposition of Normalizing Flows for Continual Anomaly Detection
DeCoFlow decomposes normalizing flow subnets into frozen bases and low-rank adapters with alignment, auxiliary layers, and tail-aware loss to achieve continual anomaly detection with zero forgetting and few added parameters.
-
GoTTA be Diverse: Rethinking Memory Policies for Test-Time Adaptation
Diversity-aware memory policies improve test-time adaptation performance most under constrained memory budgets and challenging non-i.i.d. streams.
-
Fine-Tuning Regimes Define Distinct Continual Learning Problems
The relative rankings of continual learning methods are not preserved across different fine-tuning regimes defined by trainable parameter depth.
-
Self-Evolving Cognitive Framework via Causal World Modeling for Embodied Scientific Intelligence
Proposes a self-evolving cognitive framework integrating causal world modeling, intervention-driven reasoning, and continual refinement for embodied scientific intelligence.
-
Anytime Training with Schedule-Free Spectral Optimization
SF-NorMuon is a new schedule-free spectral optimizer that closes the gap with tuned AdamW on 125M-772M parameter models across 1-8x Chinchilla horizons while providing stationarity guarantees.
-
Attention to task structure for cognitive flexibility
Task connectivity in graph-structured multi-task environments enhances generalization and stability, with stronger benefits for attention models than MLPs.
-
Out-of-Distribution Generalization in Time Series: A Survey
This is the first comprehensive survey of OOD generalization methodologies for time series, organized across data distribution, representation learning, and OOD evaluation.
-
Learning Entropy and Spatial Adaptation Dynamics of Multilayer Perceptrons for Structural Point Extraction
Spatial Learning Entropy Maps derived from MLP weight adaptations during spatial pixel prediction tasks highlight image points with high learning impact.
-
The Dynamic Gist-Based Memory Model (DGMM): A Memory-Centric Architecture for Artificial Intelligence
DGMM is proposed as an explicit graph-structured memory architecture for AI that enables persistent episodic memory, cue-based recall, and context-dependent interpretation without retraining.