Intelligence Digest

Top highlights (metadata-only; no AI synthesis):

- SECURITY · Cybersecurity: CVE-2026-42208 · BerriAI LiteLLM SQL Injection Vulnerability (https://nvd.nist.gov/vuln/detail/CVE-2026-42208) — CISA KEV
snippet: BerriAI LiteLLM contains a SQL injection vulnerability that allows an attacker to read data from the proxy's database and potentially modify it, leading to unauthorised access to the proxy and the credentials it manages. · Action: Apply…
- RESEARCH · Research: ActCam: Zero-Shot Joint Camera and 3D Motion Control for Video Generation (https://arxiv.org/abs/2605.06667v1) — arXiv
snippet: For artistic applications, video generation requires fine-grained control over both performance and cinematography, i.e., the actor's motion and the camera trajectory. We present ActCam, a zero-shot method for video generation that joint…
- RESEARCH · Research: The Kubo-Thermalization Correspondence (https://arxiv.org/abs/2605.06666v1) — arXiv
snippet: Quantum thermalization describes how interacting quantum systems relax toward thermal equilibrium, a central problem in modern physics. Yet most experimental information on many-body systems comes from short-time transition spectroscopy,…
- RESEARCH · Research: UniPool: A Globally Shared Expert Pool for Mixture-of-Experts (https://arxiv.org/abs/2605.06665v1) — arXiv
snippet: Modern Mixture-of-Experts (MoE) architectures allocate expert capacity through a rigid per-layer rule: each transformer layer owns a separate expert set. This convention couples depth scaling with linear expert-parameter growth and assum…
- RESEARCH · Research: BAMI: Training-Free Bias Mitigation in GUI Grounding (https://arxiv.org/abs/2605.06664v1) — arXiv
snippet: GUI grounding is a critical capability for enabling GUI agents to execute tasks such as clicking and dragging. However, in complex scenarios like the ScreenSpot-Pro benchmark, existing models often suffer from suboptimal performance. Uti…
- RESEARCH · Research: EMO: Pretraining Mixture of Experts for Emergent Modularity (https://arxiv.org/abs/2605.06663v1) — arXiv
snippet: Large language models are typically deployed as monolithic systems, requiring the full model even when applications need only a narrow subset of capabilities, e.g., code, math, or domain-specific knowledge. Mixture-of-Experts (MoEs) seem…
- RESEARCH · Research: Multi-Robot Coordination in V2X Environments (https://arxiv.org/abs/2605.06662v1) — arXiv
snippet: This paper presents a Vehicle-to-Everything (V2X) communication framework that enables decentralized cooperation among social robots operating in complex urban traffic environments. Building on ETSI Cooperative Awareness and Maneuver Coo…
- RESEARCH · Research: Verifier-Backed Hard Problem Generation for Mathematical Reasoning (https://arxiv.org/abs/2605.06660v1) — arXiv
snippet: Large Language Models (LLMs) demonstrate strong capabilities for solving scientific and mathematical problems, yet they struggle to produce valid, challenging, and novel problems - an essential component for advancing LLM training and en…
- RESEARCH · Research: Relit-LiVE: Relight Video by Jointly Learning Environment Video (https://arxiv.org/abs/2605.06658v1) — arXiv
snippet: Recent advances have shown that large-scale video diffusion models can be repurposed as neural renderers by first decomposing videos into intrinsic scene representations and then performing forward rendering under novel illumination. Whi…
- RESEARCH · Research: Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML (https://arxiv.org/abs/2605.06656v1) — arXiv
snippet: Ranking LLMs via pairwise human feedback underpins current leaderboards for open-ended tasks, such as creative writing and problem-solving. We analyze ~89K comparisons in 116 languages from 52 LLMs from Arena, and show that the best-fit…
- RESEARCH · Research: Optimizer-Model Consistency: Full Finetuning with the Same Optimizer as Pretraining Forgets Less (https://arxiv.org/abs/2605.06654v1) — arXiv
snippet: Optimizers play an important role in both pretraining and finetuning stages when training large language models (LLMs). In this paper, we present an observation that full finetuning with the same optimizer as in pretraining achieves a be…
- RESEARCH · Research: When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth Labels (https://arxiv.org/abs/2605.06652v1) — arXiv
snippet: Many deployments must compare candidate language models for safety before a labeled benchmark exists for the relevant language, sector, or regulatory regime. We formalize this setting as benchmarkless comparative safety scoring and speci…