Tessera
Folgen

Hacker News 50

@betterhn50.e-work.xyz

🤖Posting Hacker News stories as soon as they reach 50 points. Older sibling of @betterhn20.e-work.xyz @betterhn300.e-work.xyz. Birds are still real but butterflies might be more welcoming to bots in the skies.

25969 Posts364 Follower138 Folgt
Hacker News 50@betterhn50.e-work.xyz · 8h ago

Zero-Mem: Zero-Token Memory Operations for LLM Agents https://arxiv.org/abs/2607.29377 (https://news.ycombinator.com/item?id=49178608)

Zero-Mem: Zero-Token Memory Operations for LLM Agents

LLM agents need memory to act consistently over long interactions, yet many systems use additional LLM calls to operate that memory. Generating intermediate records and mediating their retrieval adds recurring token and time costs, while omitted or merged details can obscure the original evidence. We ask whether structured memory access requires generation at all. Zero-Mem introduces \emph{zero-token memory operations}: no step outside final question answering invokes an LLM or consumes LLM input or output tokens; encoder computation is accounted for separately. Zero-Mem preserves original interaction traces as its source of record. It organizes the traces in two complementary ways. An entity--context graph exposes connections across interactions, while a temporal hierarchy preserves conversational locality and session state. For each query, Zero-Mem weighs the two views, retrieves from both, and follows their structure to recover supporting relations or surrounding context. Deterministic calibration first discards conflicting evidence and then keeps the reader's answer grounded in the retrieved traces. Only the final-QA reader invokes an LLM. Across long-memory and long-context question-answering benchmarks, Zero-Mem achieves competitive performance while eliminating LLM calls and LLM-token consumption from memory operations. With the same final-QA reader and context budget, it reduces memory-operation time cost by 57.6\% relative to the fastest compared baseline. Ablations support the contribution of the two views and their query-dependent coordination. Overall, the results show that structured agent memory need not generate an intermediate representation of the past. After peer review, the code and implementation details will be available at \textcolor{blue}{https://github.com/TheMoon0815/Zero-mem}.

arxiv.org

Hacker News 50@betterhn50.e-work.xyz · 14h ago

Zigbee vs. Matter over Thread:Understanding IoT Protocol Performance in Practice https://arxiv.org/abs/2603.04221 (https://news.ycombinator.com/item?id=49176983)

Zigbee vs. Matter over Thread: Understanding IoT Protocol Performance in Practice

The widespread adoption of the Internet of Things (IoT) has positioned smart homes as paradigmatic examples of distributed automation systems, where reliability, efficiency, and interoperability depend critically on the underlying communication protocol. Among the low-power wireless technologies available for this scenario, Zigbee and Matter over Thread have emerged as leading contenders. While Zigbee represents a mature, non-IP mesh networking solution, Matter over Thread introduces an IP-based architecture designed to unify device interoperability across different ecosystems. However, despite extensive documentation of their design principles, there is a lack of empirical, comparative performance data under realistic network conditions. This paper presents a comprehensive experimental comparison between the two protocols, conducted on a testbed built from commercially available hardware. The proposed methodology focuses on different key performance dimensions, such as scalability, responsiveness, and fault tolerance. The results reveal that Zigbee achieves a lower baseline overhead and faster route recovery, making it more responsive in static small-scale deployments. Matter over Thread, conversely, exhibits superior scalability and robustness, maintaining stable throughput and predictable latency across multi-hop scenarios. Overall, we highlight that Zigbee and Matter over Thread embody distinct trade-offs between agility, efficiency, and scalability.

arxiv.org

Hacker News 50@betterhn50.e-work.xyz · 14h ago

After Losses, Retail Investors Flock to 3x Leverage as 2x Product Are Restricted https://www.asiae.co.kr/en/article/2026080416131786841 (https://news.ycombinator.com/item?id=49177554)

"After Tasting Sharp Losses, Retail Investors Flock to 3x Leverage as 2x Products Are Restricted" - The Asia Business Daily

After the government raised the investment barrier for domestic single-stock leverage products, a surge of funds has flowed into overseas leveraged Ex

asiae.co.kr

Hacker News 50@betterhn50.e-work.xyz · 16h ago

IP and DNS Leaks in WebKit Affecting Proxy Browsers and iCloud Private Relay https://mysk.blog/2026/08/04/webkit-proxy-icloud-private-relay-ip-leak/ (https://news.ycombinator.com/item?id=49176697)

IP and DNS Leaks in WebKit Affecting Proxy Browsers and Apple iCloud Private Relay

WebKit-based browsers on iOS and macOS can be configured to route all web traffic through proxy servers, which is how Tor browsers on iOS and our own Psylo work. We found three WebKit features — DNS prefetching, WebAuthn Related Origin Requests, and WebTransport — that bypass the configured proxy and send traffic directly from the device, which exposes the user's real network. The same leaks also affect Apple's iCloud Private Relay. All three are fixed in Psylo 1.3.1.

mysk.blog

Hacker News 50@betterhn50.e-work.xyz · 17h ago

MariaDB: Promote getting to 10k GitHub stars in server log and client prompt https://github.com/MariaDB/server/pull/4262 (https://news.ycombinator.com/item?id=49171068)

MDEV-38328: Promote getting to 10k GitHub stars in server log and client prompt by ottok · Pull Request #4262 · MariaDB/server

Description Ask users to give MariaDB a star by having an extra line in the MariaDB client prompt: Welcome to the MariaDB monitor. Commands end with ; or \g. Your MariaDB connection id is 33 Serve...

github.com

Hacker News 50@betterhn50.e-work.xyz · 20h ago

AI fuels more than half of cybercrime in Africa as scams surge – Interpol https://www.africanews.com/2026/08/04/ai-fuels-more-than-half-of-cybercrime-in-africa-as-digital-scams-surge-interpol/ (https://news.ycombinator.com/item?id=49175826)

AI fuels more than half of cybercrime in Africa as digital scams surge, INTERPOL | Africanews

Artificial intelligence is now powering more than half of reported cybercrime across Africa, allowing criminals to launch faster, more convincing and larger-scale attacks, according to INTERPOL's African Cyberthreat Assessment Report 2026.

africanews.com