CL-Bench is the first expert-validated benchmark for continual learning in frontier LLMs across six real-world domains, showing limited gains and that naive in-context learning outperforms dedicated memory systems.
Jerry Liu
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 8roles
background 1polarities
background 1representative citing papers
Doc-to-Atom decomposes documents into composable micro-LoRA adapters selected by a query router for efficient long-context QA.
A product-key parametric memory head with selective sparse updates mitigates catastrophic forgetting in generative retrieval models during sequential addition of new documents.
Different patterns of environmental change (space vs time) require different LLM update behaviors; no single family of methods—prompts, distillation, RL, or compression—handles all regimes.
DualSelect couples task and reference selection via a minimax framework with entropy-regularized scoring to preserve safety in LLM fine-tuning, reporting at least 5.10 point gains in Safety Avg. over baselines on 1B-8B models.
DualSFT derives parameter masks and data subsets as row- and column-wise aggregations of one gradient interaction matrix under first- and second-order validation-improvement approximations.
Sparse memory modules with KL-based surprising-token selection let retrofitted LLMs acquire new factual knowledge while largely preserving held-out capabilities.
ESG-adapted versions of Qwen-3-4B using LoRA and IRM outperform the base model and Llama-3/Gemma-3 baselines on generative ESG question-answering tasks.
citing papers explorer
-
Continual Learning Bench: Evaluating Frontier AI Systems in Real-World Stateful Environments
CL-Bench is the first expert-validated benchmark for continual learning in frontier LLMs across six real-world domains, showing limited gains and that naive in-context learning outperforms dedicated memory systems.
-
Doc-to-Atom: Learning to Compile and Compose Memory Atoms
Doc-to-Atom decomposes documents into composable micro-LoRA adapters selected by a query router for efficient long-context QA.
-
A Parametric Memory Head for Continual Generative Retrieval
A product-key parametric memory head with selective sparse updates mitigates catastrophic forgetting in generative retrieval models during sequential addition of new documents.
-
When Does Continual Learning Require Learning
Different patterns of environmental change (space vs time) require different LLM update behaviors; no single family of methods—prompts, distillation, RL, or compression—handles all regimes.
-
Two to Tango: Coupled Task-Reference Selection for Safe LLM Fine-tuning
DualSelect couples task and reference selection via a minimax framework with entropy-regularized scoring to preserve safety in LLM fine-tuning, reporting at least 5.10 point gains in Safety Avg. over baselines on 1B-8B models.
-
One Algorithm, Two Goals: Dual Scoring for Parameter and Data Selection in LLM Fine-Tuning
DualSFT derives parameter masks and data subsets as row- and column-wise aggregations of one gradient interaction matrix under first- and second-order validation-improvement approximations.
-
Improving Sparse Memory Finetuning
Sparse memory modules with KL-based surprising-token selection let retrofitted LLMs acquire new factual knowledge while largely preserving held-out capabilities.
-
Developing an ESG-Oriented Large Language Model through ESG Practices
ESG-adapted versions of Qwen-3-4B using LoRA and IRM outperform the base model and Llama-3/Gemma-3 baselines on generative ESG question-answering tasks.