SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2505.23195 · NeurIPS · 2025

Less is More: Unlocking Specialization of Time Series Foundation Models via Structured Pruning

Xue Wang, Lifan Zhao, Yanyan Shen, Zhaoyang Liu, Jiaji Deng

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
SJTU-DMTai/Prune-then-Finetune — 3 of 5
FunctionStatusWhere it lives
Mask Ran SJTU-DMTai/Prune-then-Finetune/src/tsfm/layers/prune_mask.py
code served (permissive licence) · get_code("92baea6ddfe32892")
accumulate_fisher Ran SJTU-DMTai/Prune-then-Finetune/src/tsfm/layers/prune_mask.py
code served (permissive licence) · get_code("efa644f09fe16714")
zero_tensor Ran SJTU-DMTai/Prune-then-Finetune/src/tsfm/layers/prune_mask.py
code served (permissive licence) · get_code("f0c01ac22ec18298")
MaskedLayer Not yet run SJTU-DMTai/Prune-then-Finetune/src/tsfm/layers/prune_mask.py
code served (permissive licence) · get_code("4ea419fd7112b240")
as_buffer_ Not yet run SJTU-DMTai/Prune-then-Finetune/src/tsfm/layers/prune_mask.py
code served (permissive licence) · get_code("50f1a80d0df9c681")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Scaling laws motivate the development of Time Series Foundation Models (TSFMs) that pre-train vast parameters and achieve remarkable zero-shot forecasting performance. Surprisingly, even after fine-tuning, TSFMs cannot consistently outperform smaller, specialized models trained on full-shot downstream data. A key question is how to realize effective adaptation of TSFMs for a target forecasting task. Through empirical studies on various TSFMs, the pre-trained models often exhibit inherent sparsity and redundancy in computation, suggesting that TSFMs have learned to activate task-relevant network substructures to accommodate diverse forecasting tasks. To preserve this valuable prior knowledge, we propose a structured pruning method to regularize the subsequent fine-tuning process by focusing it on a more relevant and compact parameter space. Extensive experiments on seven TSFMs and six benchmarks demonstrate that fine-tuning a smaller, pruned TSFM significantly improves forecasting performance compared to fine-tuning original models. This "prune-then-finetune" paradigm often enables TSFMs to achieve state-of-the-art performance and surpass strong specialized baselines. Source code is made publicly available at https://github.com/SJTU-DMTai/Prune-then-Finetune.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2505.23195")
get_code_for_paper("2505.23195")
have("2505.23195")

Connect an agent — have() is free.