The paper investigates run‑to‑run variance in semantic query engines that use large language models (LLMs) to evaluate rows. By re‑executing benchmarks (SemBench, LRO‑Bench, UDA‑Bench) and logging 1.2 million model calls, the authors find that between 0.07% and 10.2% of rows change output at temperature 0, with some tasks showing up to 24% change. They show that plan amplification can cause a large fraction of answers to differ, that many benchmark comparisons are unresolved, and that voting and calibrated cascades reduce but do not eliminate variance, leading to new reporting and design guidelines.
arXiv:2607. 04281v1 Announce Type: cross Abstract: Semantic caching reduces the latency and cost of retrieval-augmented generation (RAG) by serving cached answers to semantically similar queries, but most existing methods do not model the time-varying freshness of open-web evidence.
The paper investigates how prefix caching, a default optimization in open‑source LLM serving stacks, affects reproducibility when combined with weight quantization. Experiments on an eighty‑episode multi‑turn agentic tool‑use workload show that enabling the cache causes the agent’s trajectory to change in 36.2 % of episodes at 16‑bit precision and 75.0 % at 4‑bit precision, while disabling the cache yields perfectly reproducible runs. The study identifies specific cache‑related settings that drive run‑to‑run divergence and demonstrates that cached serving is deterministic only when the cache state is preserved, which is not the case in typical deployments.
arXiv:2609.38006v1 Announce Type: new Abstract: Running a language model locally offers advantages in privacy, latency, and cost, but local hardware fits only small models, which are less capable tha...
Every memory-based knowledge editor in the SERAC lineage depends on a scope decision: given a query, does a stored edit apply? We report that current knowledge-editing benchmarks cannot measure this d...
The paper argues that retrieval‑augmented question‑answering systems should perform semantic compilation at ingest time rather than re‑deriving meaning at query time. By building a maintained structure—incrementally updated embeddings and validated atomic claims—read operations become far cheaper, with experimental results showing higher accuracy and lower token usage compared to traditional chunk‑based retrieval. The authors present two proofs: cheaper incremental updates and superior performance on broadcast‑interview transcripts, suggesting a new systems agenda for compilation and read planning.
arXiv:2608. 09254v1 Announce Type: new Abstract: LLM analytics agents are evaluated on SQL syntax accuracy, but production failures look different: questions with two valid business definitions, questions the warehouse cannot answer, deprecated columns after a schema change, and queries that execute successfully while returning the wrong business number.
arXiv:2610. 12768v1 Announce Type: cross Abstract: Semantic query engines evaluate natural-language filters with a large language model, and every evaluation is a costly oracle call.
The paper investigates how test‑time computation can enhance language models and at what cost, introducing the SELF‑POT benchmark to evaluate this across competition mathematics, competitive programming, and agentic workflows. SELF‑POT separates candidate coverage from final accuracy, tracks correctness transitions under revision, and measures protocol completion alongside task success. Using a unified budget rule, the study compares direct inference, parallel sampling, and self‑revision across five low‑cost reasoning models, revealing that selection rules and failure handling significantly influence gains and cost savings.
arXiv:2608. 13612v1 Announce Type: new Abstract: Natural-language interfaces to enterprise data must translate underspecified requests into governed, executable behavior while controlling invalid queries, policy failures, cost, and nondeterminism.
arXiv:2609.22259v1 Announce Type: new Abstract: Context layers, curated documentation that an analytics agent fetches at query time, produce large accuracy gains on text-to-SQL benchmarks. A with/wit...
ESQ‑Bench is an Oracle‑first NL2SQL benchmark that introduces systematic complexity tiers and silent‑divergence evaluation across three enterprise schema levels. It provides six populated schemas (465 tables, 164,682 rows) on Oracle, PostgreSQL, MySQL, and SQL Server, along with 550 gold‑validated question‑query pairs and a four‑metric evaluation harness. The benchmark reveals that state‑of‑the‑art models such as GPT‑4o and Claude Sonnet 4.6 experience significant drops in execution accuracy and high silent‑divergence rates as schema complexity increases, highlighting a gap between closed‑API models and open‑weight baselines on enterprise Oracle schemas.
The paper reports that existing knowledge‑editing benchmarks cannot evaluate the scope decision—whether a stored edit applies to a query—because they are counterfactual and lack negative examples. Using the gradient‑free editor INLAY, the authors exhaustively test every router action on 1,689 queries across three datasets and find that an oracle router achieves no gain over a static policy, and abstention never wins. The authors attribute this to the structural design of the benchmarks and demonstrate that adding a missing negative condition restores some headroom and allows abstention to win.