1
ESS 推出“全球最高性能”音频编解码器 SABRE ES9292PRO
IT之家 8 月 6 日消息，音频半导体企业 ESS Technology 美国加利福尼亚州当地时间 5 日宣布推出 “全球最高性能”音频编解码器 SABRE ES9292PRO 。 SABRE ES9292PRO 基于 ESS 旗舰级 PRO 立体声 ADC / DAC 首创的 Hyperstre…
2
Temperature-driven inversion and nonlinear dynamics in ChatGPT-like AIs
发现ChatGPT类AI中温度升高反而降低熵的奇特现象，颠覆直觉的动力学研究。
arXiv:2608.00939v1 Announce Type: cross Abstract: Increasing the temperature of an ordinary many-state system increases access to a wider range of sta…
3
Measuring the Redundancy of Decoder Layers in SpeechLLMs
语音大模型解码器层存在多少冗余？这篇论文提出测量方法，为模型压缩提供新视角
arXiv:2603.05121v2 Announce Type: replace-cross Abstract: Speech Large Language Models route speech encoder representations into an LLM decoder that t…
4
Cross-Attention is Half Explanation in Speech-to-Text Models
语音转文本模型中的交叉注意力并非完整解释，揭示其局限与真正作用机制
arXiv:2509.18010v2 Announce Type: replace-cross Abstract: Cross-attention is a core mechanism in encoder-decoder architectures, widespread in many fie…
5
DECSELFMASK: Leveraging Unlabeled Text via Self-Relevance-Guided Masking for Decoder-Only Classification
自监督掩码学习突破标注瓶颈，医疗等场景下用无标签文本提升分类性能。
arXiv:2606.09466v1 Announce Type: new Abstract: Classification tasks require annotated data, which can often be expensive, time-consuming, or even unf…
6
STREAM: Stochastic Riemannian Flow Matching with Anisotropic Decoder for Digital Histopathology Image Generation
新方法STREAM结合随机黎曼流匹配与各向异性解码器，精准生成数字病理图像，破解医疗数据隐私与训练数据短缺难题。
arXiv:2606.07036v1 Announce Type: cross Abstract: Synthetic histopathology image generation addresses critical challenges in computational pathology, …
7
AV2 v1.0.0 视频编码标准发布：完整规范已公开，硬件支持可能还要等几年
AV2 v1.0.0正式发布，压缩效率提升约30%，完整规范已公开，但硬件支持仍需等待数年。
IT之家 6 月 8 日消息，开放媒体联盟（AOMedia）现已发布 AV2 v1.0.0 视频编码标准，主要提升压缩效率，并加入新场景支持，目前完整规范已经在官网公开。 IT之家注意到，AV2 的设计目标不仅是提升压缩效率， 还加强 AR / VR 内容 、 多画面传输 、 屏幕内容编码等场景支持…
8
ReverseEOL: Improving Training-free Text Embeddings via Text Reversal in Decoder-only LLMs
无需训练，仅通过反转输入文本，就能显著提升解码器LLM的文本嵌入质量。
arXiv:2606.05858v1 Announce Type: new Abstract: Recent advances in Large Language Models (LLMs) have opened new avenues for generating training-free t…
9
ResNet-34 with Lightweight Decoder for Accurate and Efficient Segmentation of Fetal Brain MRI
轻量级解码器搭配ResNet-34，在胎儿脑MRI分割中实现精度与速度的双重提升
arXiv:2606.01293v1 Announce Type: cross Abstract: Accurate segmentation of fetal brain tissues in Magnetic Resonance Imaging (MRI) is critical for ear…
10
DOA: Training-Free Decoder-Only Attention Policy for Long-Form Simultaneous Translation with SpeechLLMs
无需训练的纯解码器注意力新策略，让语音大模型长文本同声翻译更高效稳定。
arXiv:2605.31432v1 Announce Type: cross Abstract: Simultaneous speech-to-text translation (SimulST) generates translations while speech is still unfol…
11
Faithfulness Evaluation for Decoder-only LLM Attributions with Controlled Retained Information
提出控制保留信息的新方法，可靠评估Decoder-only LLM的归因忠实度，为可解释性研究提供关键视角。
arXiv:2601.03089v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly evaluated with input attribution methods, yet …
12
Show HN: Rav2d – AV2 video decoder ported from C to Rust (47K lines, 786 tests)
47K行Rust移植AV2视频解码器，786测试验证对C参考实现的字节级兼容性。
Article URL: https://github.com/stukenov/rav2d Comments URL: https://news.ycombinator.com/item?id=48247274 Points: 1 # Comments: 0