中文
相关论文

相关论文: Prompt-Activation Duality: Improving Activation St…

200 篇论文

This work introduces a new Transformer model called Cached Transformer, which uses Gated Recurrent Cached (GRC) attention to extend the self-attention mechanism with a differentiable memory cache of tokens. GRC attention enables attending…

计算机视觉与模式识别 · 计算机科学 2023-12-21 Zhaoyang Zhang , Wenqi Shao , Yixiao Ge , Xiaogang Wang , Jinwei Gu , Ping Luo

Trajectory prediction models in autonomous driving are vulnerable to perturbations from non-causal agents whose actions should not affect the ego-agent's behavior. Such perturbations can lead to incorrect predictions of other agents'…

机器人学 · 计算机科学 2026-05-19 Ehsan Ahmadi , Ray Mercurius , Soheil Alizadeh , Kasra Rezaee , Amir Rasouli

Driver's cognitive ability at a given moment is the most elusive variable in assessing driver's safety. In contrast to other physical conditions, such as short-sight, or manual disability cognitive ability is transient. Safety regulations…

人机交互 · 计算机科学 2022-04-26 Moti Salti , Yair Beery , Erez Aluf

Activation steering methods are widely used to control large language model (LLM) behavior and are often interpreted as revealing meaningful internal representations. This interpretation assumes that steering directions are identifiable and…

机器学习 · 计算机科学 2026-04-02 Sohan Venkatesh , Ashish Mahendran Kurapath

Self-attention has greatly contributed to the success of the widely used Transformer architecture by enabling learning from data with long-range dependencies. In an effort to improve performance, a gated attention model that leverages a…

机器学习 · 计算机科学 2026-02-03 Viet Nguyen , Tuan Minh Pham , Thinh Cao , Tan Dinh , Huy Nguyen , Nhat Ho , Alessandro Rinaldo

Applying steering vectors to large language models (LLMs) is an efficient and effective model alignment technique, but we lack an interpretable explanation for how it works-- specifically, what internal mechanisms steering vectors affect…

机器学习 · 计算机科学 2026-04-10 Stephen Cheng , Sarah Wiegreffe , Dinesh Manocha

Recent work on activation and latent steering has demonstrated that modifying internal representations can effectively guide large language models (LLMs) toward improved reasoning and efficiency without additional training. However, most…

机器学习 · 计算机科学 2026-01-07 Tuc Nguyen , Thai Le

Controllable generation requires language models to realize output characteristics such as reading level, politeness, and toxicity. Existing steering methods are often indirect, require access to internal activations, or depend on auxiliary…

计算与语言 · 计算机科学 2026-05-29 Hyeseon An , Shinwoo Park , Hyundong Jin , Yo-Sub Han

Controlling specific behaviors in large language models while preserving their general capabilities is a central challenge for safe and reliable artificial intelligence deployment. Current steering methods, such as vector addition and…

机器学习 · 计算机科学 2025-10-31 Hieu M. Vu , Tan M. Nguyen

Steering vectors offer a training-free mechanism for controlling reasoning behaviors in large language models, but constructing effective vectors requires identifying genuine behavioral signals in the model's hidden states. For behaviors…

计算与语言 · 计算机科学 2026-04-03 Haomin Zhuang , Hojun Yoo , Xiaonan Luo , Kehan Guo , Xiangliang Zhang

Autonomous driving in multi-agent dynamic traffic scenarios is challenging: the behaviors of road users are uncertain and are hard to model explicitly, and the ego-vehicle should apply complicated negotiation skills with them, such as…

机器人学 · 计算机科学 2022-06-22 Peide Cai , Hengli Wang , Yuxiang Sun , Ming Liu

Understanding how large audio models represent music, and using that understanding to steer generation, is both challenging and underexplored. Inspired by mechanistic interpretability in language models, where direction vectors in…

Previous studies have demonstrated the effectiveness of point-based neural models on the point cloud analysis task. However, there remains a crucial issue on producing the efficient input embedding for raw point coordinates. Moreover,…

计算机视觉与模式识别 · 计算机科学 2024-07-15 Zihao Li , Pan Gao , Kang You , Chuan Yan , Manoranjan Paul

We present an approach called Dialogue Action Tokens (DAT) that adapts language model agents to plan goal-directed dialogues. The core idea is to treat each utterance as an action, thereby converting dialogues into games where existing…

计算与语言 · 计算机科学 2024-06-19 Kenneth Li , Yiming Wang , Fernanda Viégas , Martin Wattenberg

Large language models (LLMs) can memorize and reproduce training sequences verbatim -- a tendency that undermines both generalization and privacy. Existing mitigation methods apply interventions uniformly, degrading performance on the…

机器学习 · 计算机科学 2026-02-10 Xuanqi Zhang , Haoyang Shang , Xiaoxiao Li

The recent works on a deep learning (DL)-based joint design of preamble set for the transmitters and data-aided active user detection (AUD) in the receiver has demonstrated a significant performance improvement for grant-free sparse code…

信息论 · 计算机科学 2022-09-07 Minsig Han , Ameha Tsegaye Abebe , Chung G. Kang

Attention mechanisms underpin the success of large language models (LLMs), yet their substantial computational and memory overhead poses challenges for optimizing efficiency and performance. A critical bottleneck arises as KV cache and…

计算与语言 · 计算机科学 2025-07-24 Luoyang Sun , Cheng Deng , Jiwen Jiang , Xinjian Wu , Haifeng Zhang , Lei Chen , Lionel Ni , Jun Wang

Masked diffusion language models (MDLMs) generate text via iterative masked-token denoising, enabling mask-parallel decoding and distinct controllability and efficiency tradeoffs from autoregressive LLMs. Yet, efficient representation-level…

计算与语言 · 计算机科学 2026-03-31 Adi Shnaidman , Erin Feiglin , Osher Yaari , Efrat Mentel , Amit Levi , Raz Lapid

Inference-time guided sampling steers state-of-the-art diffusion and flow models without fine-tuning by interpreting the generation process as a controllable trajectory. This provides a simple and flexible way to inject external constraints…

人工智能 · 计算机科学 2026-05-21 Xuehui Yu , Fucheng Cai , Meiyi Wang , Xiaopeng Fan , Harold Soh

Vision-Language-Action (VLA) models leverage powerful perceptual priors from web-scale Vision-Language Model (VLM) pre-training, yet they remain surprisingly brittle in practice, frequently failing at simple robotic tasks. To mitigate this,…

机器人学 · 计算机科学 2026-05-19 Miranda Muqing Miao , Subin Kim , Brandon Yang , Lyle Ungar