REVIEW 7 cited by
Fine-tuned Language Models are Continual Learners
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Recent work on large language models relies on the intuition that most natural language processing tasks can be described via natural language instructions. Language models trained on these instructions show strong zero-shot performance on several standard datasets. However, these models even though impressive still perform poorly on a wide range of tasks outside of their respective training and evaluation sets. To address this limitation, we argue that a model should be able to keep extending its knowledge and abilities, without forgetting previous skills. In spite of the limited success of Continual Learning we show that Language Models can be continual learners. We empirically investigate the reason for this success and conclude that Continual Learning emerges from self-supervision pre-training. Our resulting model Continual-T0 (CT0) is able to learn diverse new tasks, while still maintaining good performance on previous tasks, spanning remarkably through 70 datasets in total. Finally, we show that CT0 is able to combine instructions in ways it was never trained for, demonstrating some compositionality.
Forward citations
Cited by 7 Pith papers
-
Teaching a Language Model to Speak the Language of Tools
LoRA fine-tuning of BgGPT models on a bilingual Bulgarian function-calling dataset yields large gains on a self-built 120-case benchmark while keeping knowledge benchmarks stable.
-
Full-Parameter Continual Pretraining of Gemma2: Insights into Fluency and Domain Knowledge
EWC during Lithuanian continual pretraining of Gemma2 reduces catastrophic forgetting of English and improves Lithuanian fluency and accuracy on 5 of 7 knowledge benchmarks.
-
Mitigating Catastrophic Forgetting in Continual Learning through Model Growth
Growth-based pretraining (StackLLM) forgets less in continual fine-tuning, especially on reading comprehension, but reduces social bias less than standard training.
-
Enhancing Memory Recall in LLMs with Gauss-Tin: A Hybrid Instructional and Gaussian Replay Approach
Gauss-Tin, a replay method using a Gaussian mixture model with prompt-guided exemplar selection, reports positive backward transfer on the Natural Instructions benchmark versus sequential fine-tuning.
-
A Simple Baseline for Stable and Plastic Neural Networks
RDBP combines a new activation (ReLUDown) with decaying backpropagation to achieve strong stability and plasticity on Continual ImageNet with low overhead.
-
Continual Learning for Generative AI: From LLMs to MLLMs and Beyond
A survey that categorizes continual learning methods for generative models into architecture-based, regularization-based, and replay-based paradigms across four model families.
-
Unleashing the Power of Continual Learning on Non-Centralized Devices: A Survey
A review of non-centralized continual learning that taxonomizes data-, model-, and device-level methods and benchmarks twelve federated continual learning methods on six datasets.
Discussion (0). Continue with ORCID to comment.