What I learned from a month with WalShadow on ClickHouse Managed Postgres
Hidehiko ASAHI shares a month of WalShadow private-preview testing, comparing replication latency with ClickPipes and checking data correctness, schema changes, and recovery.
AI NEWS JOURNAL
Search AI news
Search headlines, summaries and articles to connect the dots.
Holistic Parallel Supervision for Expressive Speech to Speech Translation
arXiv:2611.06010v1 Announce Type: new Abstract: Expressiveness is central to speech-to-speech translation (S2ST), yet existing datasets rarely align source and target speech jointly in content, prosody, timing, and speaker identity. Consequently, current systems often rely on indirect conditioning or transfer mechanisms. To obtain this supervision, we introduce a training-data construction strategy that combines professional dubbing, which provides human-performed correspondence in content, prosody, and timing, with multilingual voice conversion to reduce speaker mismatch. We evaluate this supervision using existing expressive S2ST systems and DirectS2ST, an end-to-end method developed in this work that predicts the complete target-speech representation directly from source speech. Experiments on English-German and English-Spanish show that holistic parallel supervision improves expressive realization across all evaluated systems. DirectS2ST exhibits the largest gains in content, prosody, timing, speaker similarity, and human-rated expressiveness. To our knowledge, this work provides the first systematic investigation of holistic parallel supervision for expressive S2ST.
How Infinitas Learning uses ClickHouse Managed Postgres to process billions of events for 1.6M unique learners
Infinitas Learning runs transactions for 1.6 million learners on ClickHouse Managed Postgres and serves real-time analytics to teachers from ClickHouse Cloud, where distributed cache cut P99 latency from about 13 seconds to 410 milliseconds.
How to assess your organization’s AI readiness
AI adoption moves fast. Readiness takes longer. Assess 6 dimensions to see where your data foundation limits AI.
Oracle wants to pull stalled AI agents out of demo mode and into production
Companies want AI agents faster than they can build the AI architecture to run them. Most can put together a convincing demo. Far fewer can move agents into production without exposing data, hurting uptime or blowing their budgets. That gap is pushing database providers to rethink what a data platform for agents has to deliver. […] The post Oracle wants to pull stalled AI agents out of demo mode and into production appeared first on SiliconANGLE.
Same model, better answer: Enterprise data is what turns generic AI into a business tool
Generic AI models can write SQL, but they don’t know a company’s orders, customers or revenue definitions. Now, Oracle Database aims to give AI agents the context to move a stalled order forward. But that shift raises questions about complexity, lock-in and security as AI agents start acting on live transactions. Oracle Corp.’s answer is […] The post Same model, better answer: Enterprise data is what turns generic AI into a business tool appeared first on SiliconANGLE.
Customers are pulling data back into the Oracle database as agents demand real-time access
A shared semantic layer is fast becoming the dividing line between lakehouses built for reporting and those built for artificial intelligence. As AI agents join human analysts in querying corporate data, enterprises need platforms that make sense of information scattered across clouds, formats and rival systems. Oracle Corp. has spent the past year rebuilding its […] The post Customers are pulling data back into the Oracle database as agents demand real-time access appeared first on SiliconANGLE.
Fencing a Postgres primary that might still be alive
Promoting a standby doesn't stop the old Postgres primary from taking writes. Here's how the control plane behind ClickHouse Managed Postgres fences it on bare metal, AWS and GCP, and why the design is fail-open.
Oracle connects AI-driven testing with a new approach to database security
Frontier AI models that uncover software flaws by the hundreds are forcing database providers into larger, faster patch cycles, and real application testing is becoming the safeguard that keeps those fixes from breaking production systems. The same AI wave is also undermining security rules written into applications. Oracle Corp. has responded on two fronts, hardening […] The post Oracle connects AI-driven testing with a new approach to database security appeared first on SiliconANGLE.
CHROMA-MILL: Joint Low-Light Enhancement and White Balance under Colored Illumination
arXiv:2611.04092v1 Announce Type: new Abstract: Capturing an image in a low-light environment produces a weak, noisy sensor signal. Low-light image enhancement (LLIE) aims to restore it. Auto white balance (AWB) is also required for every image, and the same noisy signal degrades its ability to infer the scene illuminant. LLIE is typically performed first, followed by AWB, but this strategy remains largely unexamined because existing low-light datasets span a narrow range of illuminants and lack illuminant ground truth for evaluating AWB. We address this with CHROMA-MILL, a RAW dataset of 50 scenes captured with a DSLR and smartphone under four calibrated illuminants at eleven intensity levels. Because light is additive, we blend these captures in the linear RAW domain to synthesize physically valid illuminant mixtures with exact chromaticity ground truth. Benchmarking on this data, we find joint correction is best, and introduce CTI-Retinexformer, which supervises an illumination map, a chromaticity prediction, and an illumination-invariant scene embedding, outperforming the best cascade by 0.7 dB PSNR$_L$ and 1.1 $\Delta E_{00}$.
Minimal Superior Penalties: Characterising and Calibrating Decision-Aware Proper Scoring Rules
arXiv:2611.03734v1 Announce Type: new Abstract: Training a classifier means letting a validation metric decide which model is kept and when training stops. A natural requirement on that metric is superiority: every correctly classified forecast should score better than every incorrectly classified one, however close the underlying probabilities are. Neither the Brier score nor the logarithmic score satisfies this on its own. We show that superiority is governed by the separation between two endpoints -- the supremum of the score on correct predictions and the infimum on incorrect predictions -- and derive both in closed form for each score. Their difference is the infimal penalty, the greatest lower bound on the penalty needed to restore superiority. Prior work restores superiority with a penalty calibrated from the first endpoint alone; for the Brier base, our infimal penalty is smaller than that published penalty by a factor ranging from 4.00 at three classes to 2.11 at twenty. We further embed the published and infimal penalties in a one-parameter family and show that, under a single-valued tie convention, every member of the family is strictly proper. In a pre-specified two-dataset case study, mean selected-model macro-F1 rises with the penalty in the reported ten-class regimes, though not distinguishably from noise once seed/selection-mode nesting is accounted for; a fifth regime at forty-seven classes does not reproduce even this descriptive pattern. The intended no-superiority baseline is algebraically a rescaling of the family under test, so the experiment cannot attribute the observed pattern specifically to superiority. When superiority itself is the objective, any penalty strictly above the infimal value suffices; choosing a larger penalty is a separate calibration trade-off rather than a requirement for the guarantee.
What Is This Memory Useful For? Measuring Task-Conditioned Utility in Long-Term Agent Memory
arXiv:2611.03930v1 Announce Type: new Abstract: A stored experience may answer one future question and offer little help with another. We measure this task-conditioned utility rather than assigning each memory a single importance score. A task-blind writer stores a memory without seeing the later task; a frozen reader then answers that task with and without the memory, using identical filler. A cost ledger records the resources used to form, store and expose the memory. On single-call factual-recall tasks built from LongMemEval-V2 web-agent memories, the same memory helps more when it contains information needed for the answer. Its mean paired reward gain on these answer-bearing task pairs exceeds that on matched-control pairs by $+0.33$ (95% interval $+0.21$ to $+0.46$). The tested task-blind summaries and structured records yield smaller gains than Raw, including at fixed total context; a budget sweep is consistent with retention of answer-bearing content contributing to the difference. The bearing--control contrast remains positive in a new session with randomized memory position and under a second reader, although individual memory--task labels change. Generic retrieval performs above chance but misses most screen-defined answer-bearing memories. The factual-recall gains do not extend to the tested software-engineering and scientific-coding benchmarks, where even raw memories have no measurable effect. We provide a public, versioned Conditional Memory Utility Dataset of paired measurements and costs. The practical takeaway is to evaluate memory compression by the task utility it preserves alongside its costs. Task-conditioned measurements make this comparison possible; their stability must also be checked.
Structured Maps for Landmark-Based Route Instructions with LLMs
arXiv:2610.36582v1 Announce Type: new Abstract: To generate landmark-based route instructions, models must understand both map geometry and human landmark preferences. Although large language models (LLMs) show great abilities in language tasks, the best way to structure map data for these models is not yet established. This paper proposes a structured text representation for landmark selection tasks in route navigation, utilizing SVG format to encode spatial features. The structured text maintains a geometric connection with rendered map images, allowing language models to directly access spatial and semantic data. Using a new dataset created from human-written route descriptions and OpenStreetMap data, we evaluate models on two tasks: landmark selection at turns and around destinations.
The Agentic Analytics Benchmark: ranking tables and labeling failures with Jev
Jev ranks warehouse tables and labels agent failures in under a second, but adding retrieval hints leaves accuracy flat across 20 models on the Agentic Analytics Benchmark.
The new cost-based join optimizer with reordering support for non-inner joins
ClickHouse’s new exhaustive cost-based join optimizer safely reorders inner, outer, semi, and anti joins to find the cheapest valid plan.
How to put generative AI directly into a dbt pipeline with plain SQL
Google Cloud on wiring BigQuery's AI SQL functions into dbt pipelines, from sentiment scoring to LLM-generated product briefings.
How KOHO built a Claude-powered self-serve analytics tool on years of dbt governance
KOHO explains why AI self-serve analytics needs a strong data culture and a governed dbt foundation first.
Why an AI agent's revenue query was $3.5 million wrong, and how AWS fixed it with dbt Semantic Layer
A plausible AI revenue query was $3.5 million wrong. AWS explains how business context helps prevent those errors.
Why most of your dbt models aren't in a hurry, and what that's worth in Snowflake credits
Most models in a dbt DAG aren't on the critical path. Espresso AI and Authentic Insurance on turning that slack into real savings.
How Ecolab used an AI migration agent to shrink 84,000 lines of legacy SQL by 6x
Ecolab and Datafold used an AI migration agent to rebuild a decade of tangled Snowflake logic as clean, documented dbt models.