Knowledge catalogue
arXiv:2605.05704v2 Announce Type: replace-cross Abstract: Recent advances in foundation models have transformed LLMs from passive conversational systems into autonomous agents capable of reasoning and
arXiv:2605.23157v1 Announce Type: new Abstract: The attack surface of a multimodal large language model (MLLM) is language-dependent in ways that reveal the mechanistic structure of alignment failures
arXiv:2605.23904v1 Announce Type: new Abstract: Agent skills today are hand-crafted, generated one-shot, or evolved through loosely controlled self-revision, none of which behaves like a deep-learning
arXiv:2605.23035v1 Announce Type: cross Abstract: Intermediate layers of large language models (LLMs) best predict human brain responses to language, one of the most robust findings in computational n
arXiv:2412.14642v4 Announce Type: replace Abstract: Recently, Large Language Models (LLMs) have demonstrated great potential in natural language-driven molecule discovery. However, existing datasets a
arXiv:2605.23597v1 Announce Type: new Abstract: Matching person names across heterogeneous records is a core challenge in entity resolution, especially within linguistically and culturally complex env
arXiv:2605.21963v1 Announce Type: new Abstract: Long-horizon clinical simulation -- predicting how a patient's physiology evolves over years under specified interventions -- is central to chronic-dise
arXiv:2605.21541v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) remain vulnerable to transfer-based targeted attacks, where perturbations optimized on open-source surrogate
arXiv:2603.16077v3 Announce Type: replace Abstract: Masked diffusion models (MDM) exhibit superior generalization when learned using a Partial masking scheme (Prime). This approach converts tokens int
One implication of the below is that we really do mandatory preflight checks (as I argued to the Senate in 2023) asap. Anthropic is playing this one in a sober way, but a company like OpenAI might not
arXiv:2605.22297v1 Announce Type: new Abstract: Learning rate configuration is a fundamental aspect of modern deep learning. The prevailing practice of applying a uniform learning rate across all laye
Shipper’s “After Automation” is the best framing I’ve read of why AI keeps generating more work, not less. The line that stuck with me: the frame is not the framer. Models climb whatever benchmark we
The case against me below is completely intellectually dishonest, filled with lies and misrepresentations, wrong about almost literally everything it says—a textbook example of propaganda: - I didn’t
this has been my experience as well. there definitely were improvements, specifically wrt shell based computer use, but also regressions, especially in the last 3 version bumps of flicker and gerperte
arXiv:2605.22714v1 Announce Type: cross Abstract: Large language models are routinely used as automated evaluators: to review code, moderate content, or score outputs, often with many items passing th
arXiv:2603.14987v2 Announce Type: replace Abstract: Agentic AI systems increasingly act through tool-augmented, multi-step workflows whose failures (unsafe tool use, unauthorised actions, social harm)
arXiv:2506.11060v2 Announce Type: replace-cross Abstract: Large Language Model (LLM)-based coding agents have shown promising results on coding benchmarks, but their effectiveness on systems code rema
arXiv:2605.21845v1 Announce Type: new Abstract: Suicide is a leading cause of death in the United States, and understanding the circumstances that precede it requires extracting structured information
arXiv:2510.08759v2 Announce Type: replace Abstract: Understanding the capability bottlenecks of embodied multimodal large language models (MLLMs) is crucial for improving embodied agents. However, exi
Essential context on OpenAI’s Erdos result I truly miss the age of science and transparency in AI. Especially given how much money and political power and governance and scientific understanding is at
arXiv:2605.22462v1 Announce Type: new Abstract: We propose a five-stage methodology for causal feature analysis in transformer language models (probe design, feature extraction, causal validation, rob
arXiv:2605.20203v1 Announce Type: cross Abstract: As older adults increasingly use LLM-based chatbots for companionship and assistance, a safety gap is emerging. Older adults may face vulnerabilities
arXiv:2605.20704v1 Announce Type: cross Abstract: Autonomous AI agents that spawn sub-agent swarms create a safety gap: existing credential revocation mechanisms, OAuth~2.0 introspection, OCSP, and W3
I truly miss the age of science and transparency in AI. Especially given how much money and political power and governance and scientific understanding is at stake. We don’t know for example • How man
arXiv:2605.21573v1 Announce Type: new Abstract: We introduce Lens, a 3.8B-parameter T2I model that achieves performance competitive with, and in several cases surpassing, state-of-the-art models with
arXiv:2605.22177v1 Announce Type: cross Abstract: The proliferation of large language models (LLMs) and modular skills has endowed autonomous agents with increasingly powerful capabilities. Existing f
arXiv:2510.23090v2 Announce Type: replace Abstract: Recent advances have investigated the use of pretrained large language models (LLMs) for time-series forecasting by aligning numerical inputs with L
arXiv:2605.21792v1 Announce Type: new Abstract: Text-to-SQL ensembles improve over single-candidate generation by drawing multiple SQL candidates and selecting one, but their effectiveness is bounded
arXiv:2605.20368v1 Announce Type: cross Abstract: Organizations that scan documents for sensitive information face a practical problem. Cloud services require data to be sent to external infrastructur
arXiv:2605.22456v1 Announce Type: new Abstract: Cloud-hosted LLM driver agents provide useful semantic judgments, but their inference latency exceeds stepwise vehicle-control windows. Learned world mo
arXiv:2605.21413v2 Announce Type: new Abstract: As AI becomes part of everyday learning, many courses teach students to use it mainly as a productivity tool: how to prompt, search, summarize, write, c
arXiv:2605.21491v1 Announce Type: cross Abstract: As language models accelerate scientific research by automating hypothesis generation and implementation, a new bottleneck emerges: evaluating and fil
This is the most interesting paper I have read this week. The authors test a wide range of LLMs on a massive dataset of behavioural experiments, with more than 200,000 participants and nearly 26 milli
arXiv:2511.02014v2 Announce Type: replace Abstract: The detection of Protected Health Information (PHI) in medical imaging is critical for safeguarding patient privacy and ensuring compliance with reg
arXiv:2510.05094v2 Announce Type: replace Abstract: Recent video generation models can produce smooth and visually appealing clips, but they often struggle to synthesize complex dynamics with a cohere
arXiv:2605.20196v1 Announce Type: new Abstract: We investigate the hypothesis that real-data scaling laws are governed by progressive coverage of a latent predictive contribution spectrum rather than
arXiv:2605.20807v1 Announce Type: new Abstract: Subject-driven text-to-image generation still struggles to preserve high-frequency identity details such as logos, patterns, and text. Existing methods
arXiv:2512.14896v2 Announce Type: replace Abstract: In our study, we evaluated large language model (LLM) performance on pharmacy licensure-style question-answering tasks and developed an external kno
arXiv:2605.20761v1 Announce Type: new Abstract: The rapid proliferation of AI-generated text has introduced significant challenges in maintaining the integrity of digital content. Advanced generative
If this is true, using the best public estimates we have of LLM resource use, solving this Erdos problem took 0.6–6.3 kWh of electricity and about 3–31 liters of water. So that is less than three almo
arXiv:2605.20284v1 Announce Type: new Abstract: Industrial anomaly detection has been significantly advanced by Large Multimodal Models (LMMs), enabling diverse human instructions beyond detection, pa
arXiv:2506.16950v2 Announce Type: replace Abstract: Out-of-distribution (OOD) robustness is a desired property of computer vision models. Improving model robustness requires high-quality signals from
Not so fast Google Omni - @runwayml is back in the game with Aleph 2.0. Make changes and edits to videos (up to 30s) via prompts, references or restyle a frame with your favourite image editor (NB Pro
arXiv:2605.20668v1 Announce Type: new Abstract: With the advancement of AI capabilities, AI reviewers are beginning to be deployed in scientific peer review, yet their capability and credibility remai
arXiv:2605.20809v1 Announce Type: new Abstract: While Large Language Models (LLMs) demonstrate remarkable performance on zero-shot annotation tasks, they often struggle with the specialized convention
arXiv:2605.20506v1 Announce Type: cross Abstract: Humans learn social norms and behaviors from verbal feedback (e.g., a parent saying 'that was rude' or a friend explaining 'here's why that hurt'). Ye
arXiv:2605.20602v1 Announce Type: new Abstract: Successive self-training on a language model's own outputs is widely characterized as a process of flattening: diversity drops, distributions narrow, an
arXiv:2605.20749v1 Announce Type: new Abstract: Gated Linear Units (GLU) and their variants are widely adopted in modern open-source large language model architectures and consistently outperform thei
arXiv:2605.20544v1 Announce Type: cross Abstract: Vision-language models (VLMs) are used as high-level planners for embodied agents, translating natural language instructions and visual observations i
arXiv:2605.21242v1 Announce Type: new Abstract: As robot fleets become more heterogeneous, including humanoids, rovers, quadrupeds, and drones, selecting the right robot for a task becomes a core syst
arXiv:2501.02407v3 Announce Type: replace Abstract: Rapid advances in Natural Language Processing (NLP) have revolutionized many fields, including healthcare. However, these advances raise significant
arXiv:2605.19149v1 Announce Type: new Abstract: Agents operating with computer and Web use inevitably encounter errors: inaccessible webpages, missing files, local and remote misconfigurations, etc. T
arXiv:2605.19069v1 Announce Type: cross Abstract: Code-switching -- the natural alternation between two languages within a single utterance -- represents one of the most challenging and under-studied
arXiv:2605.19711v1 Announce Type: new Abstract: Automatic speech recognition (ASR) has improved substantially in recent years, yet performance remains limited for low-resource languages. Large languag
arXiv:2511.01526v2 Announce Type: replace Abstract: Multiple-choice cloze questions are commonly used to assess linguistic proficiency and comprehension. However, generating high-quality distractors r
arXiv:2601.16823v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) exhibit remarkable capabilities, yet it remains unclear to what extent these reflect sophisticated recall or genu
arXiv:2605.18999v1 Announce Type: new Abstract: Muon and related normalized optimizers decouple the choice of update direction from the choice of step scale, but their practical performance remains se
arXiv:2605.18915v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) are vulnerable to jailbreak attacks, which can elicit harmful responses from MLLMs. Many MLLMs support multi-
arXiv:2605.18854v1 Announce Type: new Abstract: Coding agents accumulate extensive context during long-running tasks, yet fixed context windows force practitioners to choose between truncation and tas
arXiv:2605.19842v1 Announce Type: new Abstract: We propose a scalable tensorization framework for neural network compression based on slice-wise feature distillation. Unlike conventional tensor decomp