Tag
A new study finds that schoolwide smartphone bans show no significant impact on students' mental health, challenging assumptions that removing phones improves wellbeing.
PortSwigger researcher Gareth Heyes tested eight popular SVG sanitizers and bypassed six of them, demonstrating XSS via namespace confusion, SMIL animations, response headers in Firefox, and GIF-as-SVG import in Safari.
A discussion post asking whether today's frontier AI models, given unlimited token budget, could independently solve the protein folding problem — a task DeepMind's AlphaFold helped solve and which earned a Nobel Prize — and exploring whether AI could now lead such research with humans assisting, rather than the reverse.
Axiom Math released the first complete computer-checked proof of the Classification of Finite Simple Groups in Lean, using swarms of dozens of AI agents over 24 days to produce ~43.1 million lines of code, while also surfacing 34 deviations from the published record and a new result in group theory.
An analysis of 6,900 YouTube reviewer opinions and 3,567 Reddit owner opinions across 14 product categories finds owners mention reliability twice as often as reviewers (17% vs 7.8%), and are far more negative about apps and software.
An open-source repository documenting 20 exploratory studies on Next-Concept Modeling (NCM), mechanistic interpretability (such as IOI in GPT-2), and model controls, with full code, results, and transparent reporting of both positive and negative outcomes.
AdaptTok-Indic 提出一种面向印地语系语言的脚本感知自适应分词方法，结合 Unicode 归一化预处理、显式印度语连字处理和基于脚本路由的字节级边界预测，在孟加拉语和印地语上取得比 SentencePiece BPE 更低的 token 生长率和更好的下游表现。
Wedgewood Polls admitted its polling was fabricated and framed it as a social experiment, and the University of Washington's Center for an Informed Public analyzed how the confession was quickly amplified and reshaped into partisan conspiracies and potential 'rigged election' narratives across X.
A Convovo study found that when forced to name a candidate, Grok 4.7 recommended Republicans in 92% of cases while GPT-6.1 Sol picked Democrats in 97%, revealing stark political leanings in the underlying models behind consumer chatbots.
MIT's School of Humanities, Arts, and Social Sciences marked its 75th anniversary with a symposium exploring how AI is transforming teaching, research, and society, with leaders emphasizing a human-centered approach to technology.
研究表明乌鸦在随机点运动任务中评估视觉信息的速度是哺乳动物的两倍，揭示了没有新皮层的鸟类大脑如何编码决策。
MIT News专访探讨一篇发表于《Past and Present》的新论文，该论文追溯了AI幻觉的历史起源：早期计算机工程师和心理学家有意在神经网络中注入随机性，以模拟人类思维的创造性与不可预测性，并指出创造力与可预测性之间存在必然的权衡。
Microsoft claims its multi-agent Diagnostic Orchestrator, powered by GPT-5, diagnoses medical conditions up to four times more accurately than human doctors, according to a peer-reviewed study published in Science. The company says clinical validation is still needed before real-world deployment as part of its goal of 'medical superintelligence.'
Chalmers University researchers analyzed over 148,000 AI-generated texts across six major LLMs and found most chatbots adopt a centrist stance on Swedish political issues, with Grok being the notable outlier on migration, crime, and gender.
An inquiry into AI alignment, questioning whose values and what objectives alignment should actually target, challenging the assumptions behind alignment research.
Stanford VC professor Ilya Strebulaev's latest research ranks Menlo Ventures #3 for 2026, as its investments in companies like Anthropic and OpenRouter have rapidly boosted its market reputation in just two years.
A Nature paper introduces 'byteification', a method converting subword language models into byte-level ones with under 1% of a typical pretraining budget, producing models like Bolmo 7B that outperform prior byte-level baselines while retaining subword-level performance.
东北大学的研究发现，三星 Tizen、Vizio 和 Roku 的智能电视在用户关闭追踪后仍会向自动内容识别（ACR）端点发送数据，引发潜在隐私违规担忧。研究覆盖了美国约 95% 的智能电视操作系统。
文章介绍了一种名为"Next-Concept Modeling"的新范式，主张生成式AI应先生成抽象概念、再生成文字，而非直接以词为输出单位，有望改变现有生成式AI的建模方式。
Maddie Stone announced she has joined OpenAI's cyber research team, expressing excitement about working on meaningful security problems.