Suchergebnisse - (endedededecoding OR decoding) (Algorithmssssics OR algorithm)

Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models von Zhang, Ce, Wan, Zifu, Kan, Zhehan, Ma, Martin Q., Stepputtis, Simon, Ramanan, Deva, Salakhutdinov, Russ, Morency, Louis-Philippe, Sycara, Katia, Xie, Yaqi

Inhaltsangabe: “… Decoding with Generative Feedback (DeGF), a novel training-free algorithm that incorporates feedback from …”
Volltext

Leveraging Small LLMs for Argument Mining in Education: Argument Component Identification, Classification, and Assessment von Favero, Lucile, Pérez-Ortiz, Juan Antonio, Käser, Tanja, Oliver, Nuria

Inhaltsangabe: “… Argument mining algorithms analyze the argumentative structure of essays, making them a valuable …”
Volltext

STAR-RIS-aided NOMA for Secured xURLLC von Lulu, Song, Di, Zhang, Shaobo, Jia, Pengcheng, Zhu, Yonghui, Li

Inhaltsangabe: “… ), which provides additional degree of freedom to enable a secure and fair decoding order and achieve a …”
Volltext

SigChord: Sniffing Wide Non-sparse Multiband Signals for Terrestrial and Non-terrestrial Wireless Networks von Peng, Jinbo, Duan, Junwen, Lin, Zheng, Yuan, Haoxuan, Gao, Yue, Chen, Zhe

Inhaltsangabe: “… on low-cost and sub-Nyquist sampling radios. We first design a deep and rule-based interleaving algorithm …”
Volltext

SpecEE: Accelerating Large Language Model Inference with Speculative Early Exiting von Xu, Jiaming, Pan, Jiayi, Zhou, Yongkang, Chen, Siming, Li, Jinhao, Lian, Yaoxiu, Wu, Junyi, Dai, Guohao

Inhaltsangabe: “… SpecEE, a fast LLM inference engine with speculative early exiting. (1) At the algorithm level, we …”
Volltext

LMDepth: Lightweight Mamba-based Monocular Depth Estimation for Real-World Deployment von Long, Jiahuan, Zhou, Xin

Inhaltsangabe: “… , existing depth estimation algorithms often struggle to effectively balance performance and computational …”
Volltext

Antenna Activation and Resource Allocation in Multi-Waveguide Pinching-Antenna Systems von Wang, Kaidi, Ding, Zhiguo, Karagiannidis, George K.

Inhaltsangabe: “… -theoretic algorithm is developed, in which the optimal decoding order is derived and incorporated. For power …”
Volltext

One Trigger Token Is Enough: A Defense Strategy for Balancing Safety and Usability in Large Language Models von Gu, Haoran, Wang, Handing, Mei, Yi, Zhang, Mengjie, Jin, Yaochu

Inhaltsangabe: “… simple yet effective defense algorithm that identifies and explicitly decodes safety trigger tokens …”
Volltext

Semi-Clairvoyant Scheduling of Speculative Decoding Requests to Minimize LLM Inference Latency von Li, Ruixiao, Chen, Fahao, Li, Peng

Inhaltsangabe: “… Speculative decoding accelerates Large Language Model (LLM) inference by employing a small …”
Volltext

An Exploratory Framework for Future SETI Applications: Detecting Generative Reactivity via Language Models von Yu, Po-Chieh

Inhaltsangabe: “… responses in language models. Instead of assuming that extraterrestrial signals must be decoded, we evaluate …”
Volltext

Towards Bridging the Reward-Generation Gap in Direct Alignment Algorithms von Xiao, Zeguan, Chen, Yun, Chen, Guanhua, Tang, Ke

Inhaltsangabe: “… Feedback (RLHF) algorithms for aligning large language models (LLMs) with human preferences. However, DAAs …”
Volltext

ABS: Enforcing Constraint Satisfaction On Generated Sequences Via Automata-Guided Beam Search von Collura, Vincenzo, Tit, Karim, Bussi, Laura, Giunchiglia, Eleonora, Cordy, Maxime

Inhaltsangabe: “… ones, and beam search is commonly used to balance exploration and fluency during decoding. While deep …”
Volltext

A Study of Large Language Models for Patient Information Extraction: Model Architecture, Fine-Tuning Strategy, and Multi-task Instruction Tuning von Peng, Cheng, Dong, Xinyu, Lyu, Mengxian, Paredes, Daniel, Zhang, Yaoyun, Wu, Yonghui

Inhaltsangabe: “… , including: (1) encoder-only or decoder-only LLMs, (2) prompt-based parameter-efficient fine-tuning (PEFT …”
Volltext

Reading Images Like Texts: Sequential Image Understanding in Vision-Language Models von Li, Yueyan, Zhao, Chenggong, Zang, Zeyuan, Yuan, Caixia, Wang, Xiaojie

Inhaltsangabe: “… an instruction-agnostic token compression algorithm based on a plug-and-play visual decoder to improve decoding …”
Volltext

STARS: Synchronous Token Alignment for Robust Supervision in Large Language Models von Quamar, Mohammad Atif, Areeb, Mohammad, Kuznetsov, Mikhail, Ozmen, Muslum Ozgur, Celik, Z. Berkay

Inhaltsangabe: “… , we introduce STARS, Synchronous Token Alignment for Robust Supervision, a decoding-time algorithm …”
Volltext

An Information-Minimal Geometry for Qubit-Efficient Optimization von Ma, Gordon, Angelakis, Dimitris G.

Inhaltsangabe: “… algorithms explore exponentially large Hilbert spaces. We recast qubit-efficient optimization as a geometric …”
Volltext

Evolving Diffusion and Flow Matching Policies for Online Reinforcement Learning von Zhang, Chubin, Wan, Zhenglin, Chen, Feng, Yang, Fuchao, Feng, Lang, Zhou, Yaxin, Yu, Xingrui, You, Yang, Tsang, Ivor, An, Bo

Inhaltsangabe: “… generation. Building on this, we introduce GoRL (Generative Online Reinforcement Learning), an algorithm …”
Volltext

Short-Context Dominance: How Much Local Context Natural Language Actually Needs? von Vakilian, Vala, Wang, Zimeng, Rawat, Ankit Singh, Thrampoulidis, Christos

Inhaltsangabe: “… with sampling strategies beyond greedy decoding. Our experiments validate that simple thresholding of the metric …”
Volltext

Hold Onto That Thought: Assessing KV Cache Compression On Reasoning von Liu, Minghui, Palnitkar, Aadi, Rabbani, Tahseen, Jae, Hyunwoo, Sang, Kyle Rui, Yao, Dixi, Shabihi, Shayan, Zhao, Fuheng, Li, Tian, Zhang, Ce, Huang, Furong, Zhang, Kunpeng

Inhaltsangabe: “… attention computations, grows linearly with context length. A suite of compression algorithms has been …”
Volltext

LoPA: Scaling dLLM Inference via Lookahead Parallel Decoding von Xu, Chenkai, Jin, Yijie, Li, Jiajun, Tu, Yi, Long, Guoping, Tu, Dandan, Song, Mingcong, Si, Hongjie, Hou, Tianqi, Yan, Junchi, Deng, Zhijie

Inhaltsangabe: “… inference. However, current confidence-driven decoding strategies are constrained by limited parallelism …”
Volltext