Related reading
Open-weight models are fast on Neon AI Gateway. Here's why
Neon AI Gateway provides fast inference for open-weight LLMs by utilizing Databricks' optimized serving stack. This stack incorporates techniques like prompt caching, continuous batching with KV-cache paging, and a TensorRT-LLM backend, leading to significant improvements in throughput and latency.
Neonneon.com6 minRoonomics: How Economics Helped Us to Build a Churn Prediction Model
Deliveroo data scientists rebuilt their restaurant churn model to prioritize precision over recall, using rolling‑window cross‑validation, SHAP interpretability, and a ranking‑based output. Offline tests show a >90% drop in flagged restaurants and an 8× precision gain, with an A/B experiment underway to measure impact on actual churn.
Deliveroodeliveroo.engineering6 minInfinite-Parameter LLMs: Generating and Adapting Weights from Live Data
The paper presents Infinite-Parameter LLMs, where a compact hypernetwork creates feed‑forward weights from live user data and updates a Bayesian latent code online, keeping the stored model size constant while effectively having infinite parameters. This design aims to improve over standard in‑context learning and retrieval by persisting knowledge in weights and freeing context space.
Hacker News front pagearxiv.org2 minpaperHN15743E2A-Bench: Benchmarking Evidence-to-Action Reliability in Financial Chart Reasoning
E2A‑Bench is a 969‑query benchmark for financial chart reasoning that evaluates vision‑language models across a full evidence‑to‑action chain using four metrics (UCR, RCI, ECI, NDR). Experiments on 20 VLMs expose hidden failures: low‑UCR models have only 6.4 % directional coverage, oracle‑aided verification cuts unsupported claims but can kill coverage, and fine‑tuning inflates BUY:SELL ratios by…
Hugging Face Daily Papersarxiv.org1 minpaperBypassing inference bottlenecks: Accelerating complex AI search with Retrieve-for-Train
Retrieve-for-Train addresses LLM inference bottlenecks in complex AI search by using offline reinforcement learning to train a lightweight diffusion model. This enables efficient, single-pass generation of diverse, property-aligned sub-queries, bypassing slow autoregressive reasoning.
Google Researchresearch.google8 minHN2When2Think: Learning Difficulty-Aware Length Control for Efficient Hybrid Reasoning Models
When2Think introduces a post‑training framework that lets a large reasoning model decide per‑instance how much reasoning depth to allocate, using difficulty‑aware reward shaping (IDAC) and verifier rewards. It cuts token usage by ~28% while boosting Pass@3 by 10% on AIME24 and reaches 40% Pass@3 on AIME25, outperforming compression and routing baselines.
Hugging Face Daily Papersarxiv.org1 minpaper
