Newsletter archive
Applied Signals Weekly: 2026-05-17
May 11, 2026 to May 17, 2026
In this week’s edition
This week’s edition surfaces where agentic systems face the sharpest pressure points: reliability under coordination, evaluation discipline, and the practical mechanics that keep multi-agent behavior stable over time.
- Databricks brings GPT-5.5 to enterprise agent workflows — Databricks uses GPT-5.5 for enterprise agent workflows after the model set a new state of the art on the OfficeQA Pro benchmark.
- EntityBench: Towards Entity-Consistent Long-Range Multi-Shot Video Generation — Multi-shot video generation extends single-shot generation to coherent visual narratives, yet maintaining consistent characters, objects, and locations across shots remains a challenge over long sequences. Existing…
- From observability to context: What’s next for Arize Phoenix — As agents start changing software, they need a way to verify their work that includes traces, evals, feedback, and APIs. This is where Phoenix goes next — not the next release, but what this product…
- FutureSim: Replaying World Events to Evaluate Adaptive Agents — AI agents are being increasingly deployed in dynamic, open-ended environments that require adapting to new information as it arrives. To efficiently measure this capability for realistic use-cases, we propose building…
- ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both — Visual reasoning, often interleaved with intermediate visual states, has emerged as a promising direction in the field. A straightforward approach is to directly generate images via unified models during reasoning, but…
- How we use Alyx to build Alyx: How to build an AI agent feedback loop — How Arize uses Alyx to debug Alyx: searching dense traces, aggregating failures, triaging dogfooding issues, and closing the AI engineering feedback loop.
- SocialReasoning-Bench: Measuring whether AI agents act in users’ best interests — Using SocialReasoning Bench, we observed a stable pattern across models—agents execute competently, but fail to consistently improve the user’s position, even with explicit instructions to optimize for user…
Deep dive
In the real world, the hardest part of scaling isn’t raw capability—it’s keeping failure modes observable and fixable as systems become more decentralized. When evaluation loops, instrumentation, and state-management drift faster than fixes, teams accumulate “invisible” debt that shows up only during deployment.
Worth watching
- evaluation and planning often showed up together—watch whether that pairing moves from discussion into repeatable engineering patterns.
- evaluation and frameworks often showed up together—watch whether that pairing moves from discussion into repeatable engineering patterns.
- evaluation and tool-use often showed up together—watch whether that pairing moves from discussion into repeatable engineering patterns.
Read the full analysis
- Evaluation loops are the new planning bottleneck (weekly trend)
- May 17, 2026 DELEGATE-52, MemEye, and EntityBench quantify long-horizon fidelity
- May 16, 2026 PREPING Memory, Tool Knowing-Doing Gap, and SPIN DAG Planning
- May 15, 2026 PolitNuggets FactNet, GraphBit DAG orchestration, FutureSim forecasting benchmark
- May 14, 2026 MAVIC boundary Bellman correction and BenchJack exploit patching
- May 13, 2026 Spectral diagnostics, Bicameral coupling, and ReVision token pruning
- May 12, 2026 CEAD enterprise capability boundaries, MemQ provenance credit, Self-Anchored Consensus
- May 11, 2026 GraphDC subgraph agents and spectral hidden-state coalition diagnostics
Subscribe
Get Applied Signals weekly newsletter in your inbox