paper-with-me

Papers

Gold-Switch: Training-Free Superposition of Slow- and Fast- Thinking LLMs

2025-10-08 · Jaeseong Lee, Dayoung Kwon, seung-won hwang arxiv

Large Reasoning Models (LRMs) excel in structured tasks by emulating deliberate human reasoning but often suffer from overthinking, degrading performance and wasting resources. One possible baseline is to deploy both LLM and LRM, then route input by predicting whether it requires reasoning and may cause overthinking. However, deploying multiple models can be costly or impractical. We propose a superposed deployment strategy with a lightweight, training-free regulation to optimize inference by switching one model on and off. Instead of routing, we selectively unlearn from LRM at inference, scaling down computation while preserving reasoning. By analyzing the cumulative energy of singular values, we identify optimal low-rank projections to adjust reasoning just right.

📄 PDF Abstract BibTeX arXiv:2510.06750

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

The Illusion of Superposition? A Principled Analysis of Latent Thinking in Language Models

2026-04-07 · Michael Rizvi-Martel, Guillaume Rabusseau, Marius Mosbach arxiv

Latent reasoning via continuous chain-of-thoughts (Latent CoT) has emerged as a promising alternative to discrete CoT reasoning. Operating in continuous space increases expressivity and has been hypothesized to enable su…

GBA: A Tuning-free Approach to Switch between Synchronous and Asynchronous Training for Recommendation Model

2022-05-23 · Wenbo Su, Yuanxing Zhang, Yufeng Cai, Kaixu Ren 외

High-concurrency asynchronous training upon parameter server (PS) architecture and high-performance synchronous training upon all-reduce (AR) architecture are the most commonly deployed distributed training modes for rec…

Recommendation Systems

Cell-free Fluid Antenna Multiple Access Networks

2025-04-29 · Tianyu Han, Yongxu Zhu, Kai-Kit Wong, Gan Zheng 외

Fluid antenna enables position reconfigurability that gives transceiver access to a high-resolution spatial signal and the ability to avoid interference through the ups and downs of fading channels. Previous studies inve…

Superposition as Lossy Compression: Measure with Sparse Autoencoders and Connect to Adversarial Vulnerability

2025-12-15 · Leonard Bereska, Zoe Tzifa-Kratira, Reza Samavi, Efstratios Gavves arxiv

Neural networks achieve remarkable performance through superposition: encoding multiple features as overlapping directions in activation space rather than dedicating individual neurons to each feature. This challenges in…

Adversarial Robustness

Superposition in Transformers: A Novel Way of Building Mixture of Experts

2024-12-31 · Ayoub Ben Chaliah, Hela Dellagi

Catastrophic forgetting remains a major challenge when adapting large language models (LLMs) to new tasks or domains. Conventional fine-tuning often overwrites existing knowledge, causing performance degradation on origi…

Mixture-of-Experts