LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Python

SpaRTA

[ACL 2026 Main] Spectral Disentanglement: Rank-Aware Task Adaptation for Rehearsal-free Continual Learning in LLMs

PythonEmerging

Stars

8

Forks

—

Contributors

1

Last push

2mo ago

Recent commits

Latest commits.

fix(model): robust FLM-Audio text-only audio placeholder shapes
933a8f6xnhyacinth2mo ago
fix(model): preserve GenerationMixin staticmethods for remote-code models
2c6ce51xnhyacinth2mo ago
fix(cl): avoid expand_size kwarg collision in generate
15ee7a8xnhyacinth2mo ago
fix(cl): retry generate after pruning unused model_kwargs
33cc817xnhyacinth2mo ago
fix(cl): drop unsupported position_ids in generate
dfeb44axnhyacinth2mo ago

feat: skip completed tasks with ckpt/test markers

f60cb54xnhyacinth2mo ago

fix(llamafactory): Patch FLM-Audio generation helpers on class

339a4a5xnhyacinth2mo ago

fix(llamafactory): Harden FLM-Audio end-to-end CL path

a37ca7fxnhyacinth2mo ago

Top contributors

Builders behind this project.