Pith. sign in

REVIEW 2 cited by

GW-MoE: Resolving Uncertainty in MoE Router with Global Workspace Theory

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

arxiv 2406.12375 v1 pith:UZHPUBPZ submitted 2024-06-18 cs.LG cs.AI

classification cs.LGcs.AI
keywords tokensgw-moeuncertainduringexpertexpertsfine-tuningglobal
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Mixture-of-Experts (MoE) has been demonstrated as an efficient method to scale up models. By dynamically and sparsely selecting activated experts, MoE can effectively reduce computational costs. Despite the success, we observe that many tokens in the MoE models have uncertain routing results. These tokens have nearly equal scores for choosing each expert, and we demonstrate that this uncertainty can lead to incorrect selections. Inspired by the Global Workspace Theory (GWT), we propose a new fine-tuning method, GW-MoE, to address this issue. The core idea is to broadcast the uncertain tokens across experts during fine-tuning. Therefore, these tokens can acquire the necessary knowledge from any expert during inference and become less sensitive to the choice. GW-MoE does not introduce additional inference overhead. We validate that GW can mitigate the uncertain problem and consistently improve in different tasks (text classification, question answering, summarization, code generation, and mathematical problem solving) and model sizes (650M and 8B parameters).

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Calling the Bluff: Detecting Ever-Shifting Harmful Chat Dialogue via Ordered Reasoning Chain Regularization

    cs.CL 2026-08 conditional novelty 6.0 of 10

    BRACE detects harmful chat dialogue by regularizing a classifier with an ordered reasoning chain of topic, indicator, severity, and type, reaching 0.934 macro F1 on the authors' 9,000-dialogue test set.

  2. Demons in the Detail: On Implementing Load Balancing Loss for Training Specialized Mixture-of-Expert Models

    cs.LG 2025-01 conditional novelty 5.0 of 10

    Global-batch load-balancing loss, computed by synchronizing expert frequencies across parallel workers, improves MoE language model perplexity and downstream scores while enabling expert domain specialization.

Pith tools