Qiushi Engine on AstaBench E2E-Bench-Hard
arXiv:2609.08196v1 Announce Type: new Abstract: This report analyzes Qiushi Engine v0.8 across all 40 test tasks in AstaBench E2E-Bench-Hard, a benchmark that requires autonomous agents to carry a res
Knowledge catalogue
arXiv:2609.08196v1 Announce Type: new Abstract: This report analyzes Qiushi Engine v0.8 across all 40 test tasks in AstaBench E2E-Bench-Hard, a benchmark that requires autonomous agents to carry a res
arXiv:2609.07290v1 Announce Type: cross Abstract: The rapid digital transformation of manufacturing, often referred to as Industry 4.0, relies on seamless interoperability between physical and softwar
arXiv:2609.07549v2 Announce Type: replace Abstract: In recent years, automatic speech recognition (ASR) has witnessed transformative advancements driven by three complementary paradigms: data scaling,
Part 4 of the same box. Part 1 was 17 -> 25-29 t/s with the expert cache PR, part 2 was 37-41 t/s after switching to UD-Q4_K_XL and stacking MTP on the cache, part 3 was the top-k fallback that was so
arXiv:2609.10092v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly act as research agents, yet their ability to track shifts in research attention is difficult to evaluate bec
arXiv:2609.08372v2 Announce Type: replace Abstract: A dense retriever encodes a question as one vector, but the question arrives one word at a time. We read 2,144 held-out headlines from Thu Vien Phap
arXiv:2609.05512v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) impose substantial energy costs during deployment, yet current compression methods apply uniform quantization across all c
Recommended read. Interaction horizon scheduling is an underexplored control problem in agentic RL This paper from the Qwen team takes a closer look at the problem. Scaling the maximum number of envir
arXiv:2609.06542v1 Announce Type: cross Abstract: The evolution of modern-day communication networks towards optical solutions with enhanced capacity and robustness is driving interest in topological
arXiv:2609.07008v1 Announce Type: new Abstract: Multi-head latent attention (MLA) exposes many logical query heads through one packed latent KV stream. This representation is memory efficient, but it
Amazon SageMaker Inference now offers prefix-aware routing, a routing strategy that sends requests sharing the same prompt prefix to the same instance so the KV cache stays warm. In benchmarks on Llam
arXiv:1907.06994v2 Announce Type: replace-cross Abstract: Mixtures of experts (MoE) are conditional mixture models in which both the mixing proportions and the component densities depend on the predic
arXiv:2609.10397v1 Announce Type: cross Abstract: Exception Related Code (ERC), which includes throw statements, conditions (if statements) that guard those throw statements, and try/catch blocks, is
arXiv:2602.06471v2 Announce Type: replace Abstract: The architectural shape of dense Transformers has remained remarkably stable: narrow-wide-narrow feed-forward networks (FFNs) consume most non-embed
arXiv:2609.10305v1 Announce Type: new Abstract: Language models under one million parameters matter for edge deployment, domain adaptation, and reproducible research, yet a two-layer LSTM or Transform
arXiv:2604.03868v2 Announce Type: replace-cross Abstract: Many safety-critical control systems operate under latent uncertainty that sensors cannot resolve at decision time. Such uncertainty, arising
arXiv:2609.07095v1 Announce Type: new Abstract: LLM assistants often produce more answers than humans can review before users see them. Most evaluations ask whether an answer is wrong, unsupported, or
arXiv:2609.08115v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) pretraining relies on an auxiliary load-balancing loss (LBL) to drive per-expert utilization toward uniformity. Post-training i
I am just sharing my config for Qwen 3.8 27b that fits on a 5060TI, what is cool about this is that you can even get vision! and a 85K context (I have 1.5gb of headroom for more context or a better qu
arXiv:2601.09721v2 Announce Type: replace-cross Abstract: Consumer artificial intelligence chatbots are now accessed by hundreds of millions of users seeking health information, yet systematic evaluat
arXiv:2609.09999v1 Announce Type: new Abstract: Terminology-aware translation asks for more than a correct translation: the output must use the exact terms a glossary prescribes. The standard recipe,
Same weights. 50 of harness search. 3x on Terminal-Bench. We optimized the OpenCode harness for GPT-OSS-20B (high reasoning) with a meta-harness: an agent that reads failures, patches the harness, and
arXiv:2609.08126v1 Announce Type: new Abstract: We study scheming in LLM agents, in which agents covertly pursue misaligned goals. Our focus is to understand how scheming arises from the interaction o
arXiv:2609.05505v1 Announce Type: new Abstract: Systematic reviews require sustained human judgment across thousands of records, yet existing evaluations of large language models (LLMs) typically exam
arXiv:2609.06192v1 Announce Type: new Abstract: Scientific coding agents produce interdependent code, results, figures, and claims, yet evaluating final outputs alone does not establish whether their
arXiv:2609.07559v1 Announce Type: new Abstract: How do you validate a cheap, deterministic proxy for an oracle that is expensive, rate-limited, and non-stationary? We present a protocol built on adver
arXiv:2609.06226v1 Announce Type: cross Abstract: Large language models (LLMs) have recently made substantial progress in formal proof generation, yet presenting distinctive challenges in cryptographi
arXiv:2609.06149v1 Announce Type: cross Abstract: Script-based malware remains a prevalent attack technique. These scripts often contain indicators of compromise (IOCs) that provide actionable threat
arXiv:2609.09672v1 Announce Type: new Abstract: The rapid advancement of audio and multimodal large language models has unlocked transformative speech understanding capabilities, yet evaluation framew
arXiv:2609.10464v1 Announce Type: cross Abstract: Joint-Embedding Predictive Architecture (JEPA) world models learn a compact latent representation of the world that supports prediction and planning,
arXiv:2609.06647v1 Announce Type: new Abstract: Although professional workflows leverage large language models widely, the interpretation for auditing unconstrained free-text generation is usually int
arXiv:2609.10356v1 Announce Type: new Abstract: Long-term change understanding from images of the same place revisited over time is a challenging task with applications in map maintenance and urban in
arXiv:2609.06540v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in safety-critical applications, yet jailbreak attacks can conceal harmful intent through role-
arXiv:2609.07344v1 Announce Type: cross Abstract: Large language model (LLM) based agents are increasingly applied to cybersecurity tasks such as vulnerability discovery and automated penetration test
arXiv:2609.09264v1 Announce Type: new Abstract: Leading benchmarks for formal theorem proving with large language models are small collections drawn from competition math, such as the IMO and Putnam,
arXiv:2609.10276v1 Announce Type: cross Abstract: This paper introduces an evolutionary computation approach for consensus in structural Bayesian Network (BN) fusion under the constraint of limited tr
arXiv:2609.08236v1 Announce Type: new Abstract: Automatic safety judges -- systems such as Llama Guard or a GPT-4o grading prompt that decide whether a model's reply is harmful -- produce the numbers
arXiv:2609.09233v1 Announce Type: cross Abstract: How can language model agents effectively leverage libraries of reusable knowledge to solve long-horizon tasks? Recent work has increasingly focused o
arXiv:2609.09295v1 Announce Type: cross Abstract: The fixed-charge network flow problem (FCNFP) couples continuous flow allocation with discrete arc-activation decisions, making it a canonical but com
🌐 Supporting open source. Expanding deployment options. We’ll work closely with the open-source community on V4.1-Flash inference support and explore more deployment options. Planning a large-scale de
arXiv:2609.08149v1 Announce Type: new Abstract: SWE-Bench Pro has emerged as a standard benchmark for evaluating software engineering agents on challenging repository-level tasks. However, our analysi
arXiv:2609.06229v1 Announce Type: cross Abstract: Vulnerability discovery is becoming an important ability of large language model (LLM) agents: agents that silently miss real defects leave critical s
arXiv:2609.09349v1 Announce Type: new Abstract: Modern LLMs demonstrate impressive multilingual performance, yet standard benchmarks primarily reward selecting correct answers rather than evaluating g
arXiv:2609.09772v1 Announce Type: new Abstract: SymbolicLight V2 combines sparse event computation with continuous-state processing in a hybrid neuromorphic language architecture. Extending V1's spike
arXiv:2609.09552v1 Announce Type: new Abstract: In this paper, we solve two fundamental obstacles in applying LLMs to critical domains: token efficiency and faithfulness. To address both constraints j
arXiv:2609.09258v1 Announce Type: cross Abstract: We present a method for recovering the moral graph of a causal DAG from a probability distribution over discrete variables, using fully connected tens
arXiv:2507.22076v2 Announce Type: replace Abstract: Text-to-image (T2I) generation models have made significant strides but still struggle with prompt sensitivity: even minor changes in prompt wording
arXiv:2609.09048v1 Announce Type: cross Abstract: Whether a language model looks demographically biased can depend on how the audit asks its question. A charitable-aid benchmark reports that the same
I initially saw CED as just an efficiency improvement, but the more I read about it, the more it feels like an inference architecture leap. The encoder/decoder split has some pretty interesting implic
arXiv:2609.05514v1 Announce Type: new Abstract: Large Language Model (LLM)-based agents are increasingly used as proxies for human participants in social science research, yet it remains unclear wheth
arXiv:2510.08831v2 Announce Type: replace Abstract: As AI writing tools become widespread, we need to understand how both humans and machines evaluate literary style, a domain where objective standard
arXiv:2609.09368v1 Announce Type: new Abstract: We present the Living Library, an end-to-end framework for transforming fragmented digital archives into governed, conversational, in-person exhibit exp
The new DeepSeek V4.1 Flash model is mindblowing - back on top of the open-source model leaderboard and extremely cheap. It has a lot of very smart ways to be efficient and highly capable so I made a
arXiv:2609.08589v1 Announce Type: cross Abstract: Recent large language models can emit task-progress signals that agent frameworks use to decide whether a task should continue or stop, yet whether a
This is a brilliant paper. It's of the cleanest long-context agent designs I have seen in the past couple of months. Sequential memory agents read chunks one after another while maintaining a compact
This model is insane at landing pages. I asked DeepSeek V4.1 Flash & Claude Fable 5 to build me a landing page for a movie theater. Fable cost $1.21 while V4.1 Flash cost 2.6 cents, making it more tha
arXiv:2609.07591v1 Announce Type: cross Abstract: Foundation models for ground states in spin-1/2 systems are a promising method for problems ranging from quantum chemistry to identifying new phase di
arXiv:2609.07244v1 Announce Type: cross Abstract: Anomaly-based intrusion detection systems in industrial control systems (ICS) and operational technology (OT) environments are increasingly required t
arXiv:2609.09561v1 Announce Type: new Abstract: Surveys remain the primary way researchers grasp the lineage of methods within an AI subfield, but they scale poorly against the current rate of publica
arXiv:2609.10315v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has advanced language-model reasoning in domains such as mathematics and code, where objective a