微信内可能无法直接打开本站。请点右上角 ··· → 在浏览器打开，或复制链接后用系统浏览器访问。
XMT
信流 · 上滑连读 · 来源可核
豆包推出开学季学生优惠，送大学生3个月免费订阅权益
近日，豆包推出开学季学生优惠活动，面向在校大学生提供3个月免费订阅权益，并同步上线开学季技能专区，辅助学生高效学习，更好地完成新学期的各类学习任务。据了解，豆包此次开学季活动涵盖多项学生优惠。…
- 在校大学生完成学生认证后，即可免费领取3个月豆包订阅；参与官方活动，还能领取额度重置机会等更多权益。
- 具体权益领取上，在最新版的豆包电脑版找到“开学免费送 3 个月”入口，打开活动页按提示完成学生身份认证，通过即可领取相关权益。
- 与此同时，豆包还面向学生群体推出了开学季技能专区，丰富的技能涵盖新生报到、课程汇报PPT、社团推荐、秋招求职等多个实用场景。
RSS 官方收录 · 可信分层展示
Cadence滕晋庆：AI有望让芯片设计时间减半，但资深工程师仍是核心竞争力
OpenAI公布了首款自研推理芯片Jalapeño的首批实测结果引发了巨大的关注，比性能更值得关注的是，AI直接参与了这颗芯片的开发。根据OpenAI的披露，AI被用于探索不同实现方案，缩短设计、测量和验证循环，并优化芯片中的部分算术电路。…
- Jalapeño从初始设计到tapeout只用了9个月。
- 这并非“AI独立设计了一颗完整芯片”，Jalapeño由OpenAI与Broadcom共同开发。
- 这个“不完整”，恰好把一个被讨论多年的问题推到了台前：从AI参与芯片设计，到AI独立完成一颗芯片，中间还隔着什么？
RSS 官方收录 · 可信分层展示
AI Alignment through a Game-theoretic Lens: A Survey
arXiv:2608.…
- 27910v1 Announce Type: new Abstract: As large language models and incr…
- Existing alignment methods, while effective in improving helpfulness, …
- This survey reviews AI alignment through a game-theoretic lens.
RSS 官方收录 · 可信分层展示
Rubric-to-Code Credit Assignment for Reinforcement Learning
arXiv:2608.…
- 27906v1 Announce Type: new Abstract: Interactive web application gener…
- Unlike conventional code generation, application quality depends on mu…
- Standard GRPO collapses these structured outcomes into a single sequen…
RSS 官方收录 · 可信分层展示
Resource Constraints and Performance in Agentic AI Systems
arXiv:2608.…
- 27886v1 Announce Type: new Abstract: Progress toward more autonomous A…
- These mechanisms shape both task capability and operational burden.
- We compare OpenClaw and NanoBot as complete agentic systems using a pa…
RSS 官方收录 · 可信分层展示
HyQuant: Hybrid-Precision Quantization for LLM Attention
arXiv:2608.27875v1 Announce Type: new Abstract: Quantization has been widely adopted in LLM training and inference to reduce cost and improve efficiency.…
- However, low-bit quantization of the \emph{attention} module often int…
- Existing methods mainly rely on smoothing techniques to handle outlier…
- Specifically, we propose \textbf{HyQuant}, an efficient hybrid quantiz…
RSS 官方收录 · 可信分层展示
See, Hypothesize, Validate: Multimodal Agentic Framework for Discovering Governing PDEs
arXiv:2608.27869v1 Announce Type: new Abstract: Discovering governing partial differential equations (PDEs) from observational data remains a core challenge across the sciences.…
- Existing sparse-regression, symbolic-regression, and LLM-based approac…
- We introduce \textbf{MAGE} (\textbf{M}ultimodal \textbf{A}gentic \text…
- Four role-specialized agents collaborate: a \textit{Differential Obser…
RSS 官方收录 · 可信分层展示
CoRe-MoE: Compact Reusable MoE for Continual Multimodal Instruction Tuning
arXiv:2608.…
- 27867v1 Announce Type: new Abstract: Continual multimodal instruction …
- LoRA-MoE provides a promising solution by introducing expert-based cap…
- This raises a natural question: is full expert expansion necessary for…
RSS 官方收录 · 可信分层展示
From Uncertainty to Clinical Risk: Severity-Aware Conformal Planning for Interactive Medical Diagnosis
arXiv:2608.…
- 27847v1 Announce Type: new Abstract: Interactive medical diagnosis dyn…
- Existing methods commonly guide information acquisition with predictiv…
- To address these limitations, we propose Severity-Aware Conformal Clin…
RSS 官方收录 · 可信分层展示
SpikeOPD: Stable On-Policy Distillation for Autoregressive Spiking Language Models
arXiv:2608.…
- 27857v1 Announce Type: new Abstract: Spiking neural networks (SNNs) of…
- A practical alternative is ANN-to-SNN migration through knowledge dist…
- Existing migration approaches distill on fixed corpus prefixes, wherea…
RSS 官方收录 · 可信分层展示
An Empirical Evaluation of Cross-City POI Recommendation on a Large-Scale Benchmark
arXiv:2608.…
- 27840v1 Announce Type: new Abstract: Cross-city point-of-interest (POI…
- Using the recently proposed large-scale benchmark Trip World, we empir…
- Our evaluation surfaces three bottlenecks of representative state-of-t…
RSS 官方收录 · 可信分层展示
KLOD: Locality-Preserving Knowledge Editing via Non-Target Distribution Preservation
arXiv:2608.…
- 27839v1 Announce Type: new Abstract: Fine-tuning-based knowledge editi…
- In sequential editing, such unconstrained redistribution can accumulat…
- We propose KLOD, a bounded and distribution-preserving objective for f…
RSS 官方收录 · 可信分层展示