Knowledge catalogue
arXiv:2608.23740v1 Announce Type: new Abstract: Concurrent multi-agent coding promises division of labor across modules, robustness through redundancy, and parallel exploration at the natural granular
arXiv:2608.24076v1 Announce Type: new Abstract: Evaluation of agentic information retrieval remains limited to scripted interactions with uniform users, missing both natural personality diversity and
arXiv:2608.24114v1 Announce Type: new Abstract: Training multi-turn LLM agents with reinforcement learning typically relies on trajectory-level rewards, which assign a uniform advantage to every step
arXiv:2608.23642v1 Announce Type: new Abstract: AI agents pose significant risks as they are granted increasing autonomy. A commonly proposed solution is human oversight and keeping a ''human in the l
Kate Clark / Wall Street Journal: AI assistant Instinct is raising a 250M Series B co-led by Index Ventures and Benchmark at a 2.5B valuation, bringing its total raised to 350M — Buzzy startup Instinc
arXiv:2608.23875v1 Announce Type: new Abstract: Artificial Intelligence (AI) algorithms frequently learn creative and unexpected solutions, surprising even expert researchers who develop and study the
Puzzles and games have been central to AI development since the very beginning. Just as we humans like to test our smarts with crosswords or logic puzzles, developers can test how far models have adva
arXiv:2608.24046v1 Announce Type: new Abstract: When an AI algorithm makes decisions that affect more than one person, aligning it becomes a problem of social choice: how should people's divergent pre
Luz Ding / Bloomberg: Alibaba releases Qwen3.8-Flash, an open-weight, 125B-parameter model built on its next-gen Qwen 4 architecture, saying it rivals Opus 4.6 and V4-Flash — Alibaba Group Holding Ltd
Ollama’s latest release retains all existing functionalities while adding a built‑in web‑search capability. Users can now run research tasks directly through open models that automatically query the i
arXiv:2608.24051v1 Announce Type: new Abstract: Can a sequence model remain competitive with only a few thousand parameters and an explicitly auditable prediction interface? We introduce ALPHABET, a c
Mahadev Konar co‑hosted a 2‑hour live broadcast titled “Amazing deep dive on K3!!” with Together Compute. The session explored scaling the Kimi K3 language model for inference, detailing deployment st
Amazing! 🥳 Thanks @vllm_project for getting Qwen3.8-Flash-Next running on NVIDIA and AMD from day 0. Qwen3.8-Flash-Next from @Alibaba_Qwen has day-0 support in vLLM, verified on NVIDIA and AMD GPUs. 🎉
Despite rapid advances in artificial intelligence, the enterprise world is still dealing with a data pipeline problem. More than 80% of enterprise data is unstructured, and 99% of this data is dark to
America is LLM-pilled. China is World-Model-Pilled. China’s booming ultra-short drama industry is being reshaped by the rapid rise of AI-generated video. In Zhengzhou, one of the country’s major produ
arXiv:2608.24175v1 Announce Type: new Abstract: Iterated Function Systems (IFS) generate self-similar fractals from a few contractive affine maps. The forward map from parameters to images is computat
arXiv:2601.19509v2 Announce Type: replace Abstract: In loosely coupled strapdown inertial navigation system/Doppler velocity log (SINS/DVL) integration, the body-frame DVL velocity is projected into t
arXiv:2608.24127v1 Announce Type: cross Abstract: Telephone fraud is pervasive and costly, but its inner workings are rarely observed at scale. We analyze a complete corpus of 10,211 inbound scam and
Another special guest is hitting the stage at #FullyConnected26! @jerryjliu0 built @llama_index, the framework thousands of teams use to connect LLMs to their data. Come see Jerry and many more Sept 2
Frederic Lardinois / The New Stack: Anthropic gives Claude Cowork its own built-in browser on the desktop app separate from users' day-to-day browser, rolling out to paying subscribers — Anthropic add
Upgraded to a 5070 Ti so I could run Qwen 3.8 27B, which works perfectly, but didn't want to let the old 4070 Ti go to waste. The cards would touch if I put them both in the PC and I knew the heat wou
arXiv:2602.03702v2 Announce Type: replace-cross Abstract: Large language models are increasingly trained in continual or open-ended settings, where the total training horizon is not known in advance.
API is live on QwenCloud: https://www.qwencloud.com/models/qwen3.8-flash 🙌Let's build something! ⚡Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight!
arXiv:2608.23816v1 Announce Type: new Abstract: Quantized fine-tuning (QLoRA) saves memory but not time. It dequantizes every 4-bit weight on the fly, so it trains more slowly than fp16 LoRA. We prese
arXiv:2608.24099v1 Announce Type: new Abstract: GUI agents often encounter dynamic anomalies when deployed on Android devices, from unexpected pop-ups to action misuse, yet existing benchmarks lack sy
The news about Qwen 3.8 Flash Next is the first I'm reading about n-gram tables. I may be completely misunderstanding how they work but it seems they could open the door for 1T+ parameter models to be
arXiv:2512.23245v3 Announce Type: replace Abstract: Recent text-to-image diffusion models have significantly improved visual quality and text alignment. However, generating a sequence of images while
arXiv:2506.02825v3 Announce Type: replace-cross Abstract: We present the OmniMatch algorithm for seeded multiple graph matching. In the setting of d-dimensional Random Dot Product Graphs (RDPG), we pr
At a 1M-token context length, QSA’s attention kernel is up to 7.6× faster in prefill and 4.9× faster in decode. With a 90% prefix-cache hit rate, Qwen3.8-Flash-Next delivers 8.6× the prefill throughpu
arXiv:2608.23640v1 Announce Type: new Abstract: When a large language model (LLM) is asked to write a person's life, how much of what it writes actually happened? We present a scene-level case-study a
arXiv:2510.21536v5 Announce Type: replace-cross Abstract: Free-space segmentation is essential for autonomous robots to identify drivable regions and navigate safely across indoor, outdoor, and road-s
arXiv:2608.23670v1 Announce Type: new Abstract: LLM-based agents execute multi-step tasks, but their behavioral structure remains opaque: long unstructured traces resist the safety auditing and runtim
arXiv:2608.23842v1 Announce Type: cross Abstract: DevOps programming (e.g., using CLI/API scripts or IaC frameworks) is key to cloud infrastructure management. Unlike traditional programming tasks, De
arXiv:2608.24807v1 Announce Type: cross Abstract: Model cards are structured documents that summarize key information about machine learning models to improve transparency, usability, and accountabili
arXiv:2608.23691v1 Announce Type: new Abstract: We study autonomous mathematical discovery in the Station, an open-world multi-agent environment in which AI agents from different model families pursue
arXiv:2608.24364v1 Announce Type: new Abstract: Self-supervised pretraining enables transferable representations for medical imaging, yet most CT encoders remain biased toward coarse semantic understa
ggml-meta: propagate buffer usage and call init on the new tensors (#27586) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/43044002 macOS/iOS: macOS Apple
ggml-metal: add chunked SSD MMA for Mamba-2 prefill optimization (#26647) metal: WIP chunked SSD SSM_SCAN kernels for multi-token prefill metal: drop scalar SSD path; MMA + sequential tail drop WIP ss
cuda: unblock mmq for MoE on sm_60 (#26264) cuda: unblock mmq for MoE on sm_60 cuda: duplicate mmq-config-pascal for dp4a and older cuda: reduce occupancy on non-dp4a pascal for Q2_K, Q4_K, Q5_K, Q6_K
ci: Clean up UI builds from releases (#27706) ci : inline UI version resolution into ui-build.yml ci : build UI once and reuse the artifact in release jobs Server jobs now extract the ui-build artifac
Implemented vulkan cross_entropy_loss and cross_entropy_loss_back (#27216) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/43166320 macOS/iOS: macOS Apple S
vulkan: warptiles currently assume warp sizes <= 64, clamp to work around larger warps (#27726) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/43173045 mac
llama: add token ID tracking to KV cell (#27762) kv: track token id rm get_prev_tokens, move it to the main pr nits add get_prev_tokens Website: https://llama.app Attestations: https://github.com/ggml
arXiv:2608.24627v1 Announce Type: new Abstract: We study adversarial bandit maximization of monotone submodular functions under a matroid constraint. For a rank-k matroid on n elements, we give a rand
arXiv:2603.05335v3 Announce Type: replace-cross Abstract: Modern predictive systems combine predictors, sequential monitors, prediction sets, and online strategies, each with a different certificate o
arXiv:2608.24219v1 Announce Type: new Abstract: Facial attractiveness has been linked to statistical regularities such as symmetry and averageness, suggesting that beauty may depend on the ease with w
arXiv:2608.24858v1 Announce Type: new Abstract: Marginalized importance weighting evaluates a target policy by reweighting offline state-action samples with its discounted occupancy ratio, characteriz
arXiv:2608.23898v1 Announce Type: new Abstract: We introduce BenchBench-Protocol, a benchmark for large language models of 149 protocol-modification tasks recovered from modifications that scientists
arXiv:2608.24314v1 Announce Type: new Abstract: Evaluating conversational voice agents at scale re- quires reliable assessment methods that capture both observ- able interaction quality and the contex
Hi all! I've been building a local agent harness, but before writing the code, I wanted to get some intuition around whether (and how) structuring agent execution as a graph helps vs. standard tool-ca
arXiv:2608.24258v1 Announce Type: new Abstract: AI systems are increasingly evaluated for legally accountable settings, where correct outputs must also be justifiable against an applicable legal stand
arXiv:2608.24024v1 Announce Type: new Abstract: Confidence-based voting aggregates parallel LLM rollouts by weighting each with internal signals such as token log probabilities, and has been actively
arXiv:2608.23653v1 Announce Type: cross Abstract: AI agents are increasingly used for simulation-driven engineering. Physical system modeling presents different requirements from general-purpose code
arXiv:2603.17170v2 Announce Type: replace-cross Abstract: AI agents increasingly execute users' natural-language (NL) tasks by calling Web services, yet today's Web authorizes these calls through OAut
arXiv:2608.24482v1 Announce Type: cross Abstract: Mechanistic Localization bridges mechanistic interpretability and post-training optimization by isolating critical parameters via interpretative appro
arXiv:2608.23858v1 Announce Type: cross Abstract: The Agent Payments Protocol (AP2), introduced by Google, enables large language model (LLM)-driven shopping agents to authorize and execute payments o
arXiv:2608.24762v1 Announce Type: cross Abstract: Many disentanglement methods represent generative factors using Euclidean product coordinates, although the underlying factor spaces may wrap, collaps
Big news: GLM-5.3-Flash by @Zai_org has landed around #5 in the Code Arena: WebDev (#2 among open models) scoring 1634 (AutoEval). Priced at 0.15/0.5 Mtoken, it reshapes the Pareto Frontier! For compa
Big shoutout to @perplexity_ai for properly benchmarking document understanding in their Portable Computer release 🔥 They took a subset of our ParseBench benchmark (https://www.parsebench.ai/) and mea
Big thanks to @sgl_project for the day-0 support! 🙌 Qwen3.8-Flash-Next is ready to deploy with SGLang today. Congrats to @Alibaba_Qwen on launching Qwen3.8-Flash! SGLang is proud to be a day-0 partner