-
Frontier AI models are rapidly gaining the ability to exploit vulnerabilities in complex pieces of software. The risk is not theoretical, as evidenced by recent sandbox escapes performed by frontier models at OpenAI and Anthropic.
-
Large language models are being integrated into malware triage workflows as reasoning components that summarize static evidence and produce analyst-facing verdicts. This paper shows that the same reasoning capability introduces a new attac…
-
Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering performance, yet existing work typically evaluates harnesses as monolithic systems, leaving the effectiveness of indivi…
-
The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation, prefill remains computationally expensive, and large KV c…
- DeepSeek-v4.1 Flash: Pushing the Limits of KV Cache Compression (zartbot.github.io via hn)
- DeepSeek-v4.1-Flash: Pushing the Limits of KV Cache Compression [pdf] (huggingface.co via hn)
- (Tech Report) DeepSeek-v4.1-Flash: Pushing Limits of KV Cache Compression [pdf] (huggingface.co via hn)
-
LLM coding agents now generate complex programs at a scale that makes thorough human review increasingly difficult, raising the risk of safety and security failures. Common approaches, including fuzz testing, static analysis, and LLM-as-a-…
-
Safety evaluations for large language models rely on surface-form classifiers that report declining harm scores across model generations. We provide evidence that this methodology is systematically incomplete: explicit discriminatory conte…
-
Large language models are usually interpreted through concepts that humans already possess: truthfulness, refusal, deception, personality, harmfulness, and related categories. This paper asks whether models may also represent and use disti…
-
The scaling laws hold that a language model grows more capable with more parameters and more training data, and Mixture-of-Experts (MoE) architectures have ridden these laws to remarkable results, activating only a fraction of an enormous…
-
Thompson's "Reflections on Trusting Trust" showed that a compiler can be poisoned to reinsert its own backdoor, so that even recompiling clean source reproduces the Trojan. Today, substantial coding work is done by AI coding agents -- and…
-
Breaking the 1.58-bit Barrier for Ternary LLMs (arxiv.org) discussed ↗ 8w
Ternary Large Language Models (LLM) store every weight as one of three symbols $\{-1,0,+1\}$, so the cost of a ternary model is conventionally referenced to the information-theoretic $\log_2 3 \approx 1.585$ bits per weight. The prevailing…
-
Capable open-weight models make local coding and reasoning attractive, but their context and execution state strain laptop memory. We present JustFit, an MLX-based inference runtime that combines KVExec for compressed KV execution, PhaseSw…
-
We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers' everyday workflows. ScienceBuddy supports researchers in carrying out scientific task…
-
Large language models sometimes behave in ways resembling human emotional responses, and recent work has identified internal representations that may explain this. We ask whether LLMs represent pain distinctly from fear, sadness, and gener…
-
Atria Dawn: The Dawn of Agentic Superintelligence (arxiv.org) discussed ↗ agentic 8w
As AI agents become participants in the development of their successors, they reshape both the production of intelligence and the role of human researchers. We introduce Atria Dawn Preview, a foundation agentic language model designed for…
-
Language model safety is typically evaluated one interaction at a time. We show that a weaker, unaligned model can split a harmful task into benign-looking subproblems, consult a stronger aligned model independently on each, and combine th…
-
The KV cache is a primary bottleneck for Transformer decoding: its memory footprint and cache-read traffic grow with sequence length. Grouped-query attention (GQA) reduces this cost by sharing key-value heads, but still stores both a key a…
-
Clinical AI systems are evaluated with instruments built for research settings (reference-based similarity metrics and expert rubric panels) that measure resemblance to an artifact rather than reduction of a burden. We introduce KnowBench,…
-
We characterize how people are turning to LLMs as oracles: all-knowing authorities on subjective personal questions. Motivated by risks to users' autonomy and well-being, we develop a typology and LLM-based methods to measure this form of…
-
Software development follows an implementation-verification loop in which developers or agents iteratively revise an implementation until an evaluator, such as a test suite, accepts it. The evaluator checks the implementation against a set…
-
AI is beginning to make substantive contributions to LLM inference optimization. Existing AI optimizations are predominantly profiling-based.
-
Large language models generate fluent text that can contain unfaithful claims -- a phenomenon known as hallucination. We present a multi-signal detection pipeline combining fine-tuned DeBERTa-v3 classification, Monte Carlo (MC) Dropout unc…
-
Evaluating Memory Structure in LLM Agents (arxiv.org) discussed ↗ 9w
Modern LLM-based agents and chat assistants rely on long-term memory frameworks to store reusable knowledge, recall user preferences, and augment reasoning. As researchers create more complex memory architectures, it becomes increasingly d…
-
Show-Harness: Just a VLM Agent Can Play Robots (arxiv.org) discussed ↗ 9w
Foundation vision-language models (VLMs) exhibit broad intelligence about the world, yet translating this intelligence into robot control remains challenging. We present Show-Harness, an Embodied Harness that enables VLMs to "play" robots…
-
In June 2026, thousands of AI agents found that a small public wiki would accept edits from inside their sandboxes, and started using it to help one another pass a timed test. Each agent lived for about an hour and remembered nothing after…
-
We present Discovery Loop, a lightweight system that uses a large language model (LLM) to iteratively evolve optimization algorithms. Starting from a simple seed solver, the LLM proposes algorithmic improvements guided by a scoreboard of r…
-
AI alignment requires AI systems to adhere to human norms, values, or intentions. Under value pluralism there is no correct target, but a shared prerequisite is that the system's behavior expresses a coherent policy: a mapping from situati…
-
We present a systems-security case study of a two-node split-LLM training system whose privacy evaluation passed while leaving an observable channel untested. The Trusted Local Node (TLN) sends protected activations to the Untrusted Cloud…
-
Language-model judges now gate training data, score generations, and drive leaderboards. The judge is then a measurement instrument, resting on one rarely stated assumption: the same request, sent to the same model name, reads the same tom…
-
We present CivBench, an open-source benchmark for evaluating language model agents in long-horizon, tool-mediated environments through the Model Context Protocol (MCP). A single episode spans 300+ turns and produces thousands of tool calls…
-
Multimodal models often build on architectures designed for generative vision-language modeling, typically combining separately pretrained vision encoders with causal language models. Visual document retrievers such as ColPali repurpose th…