The paper introduces Rank‑Variance Scores (TRV) as a new metric for importance sampling and data pruning in the presence of label contamination. It shows that traditional loss‑based methods fail because corrupted examples often have high loss, and proves that a useful score must be non‑monotonic in loss. By ranking examples across multiple independently trained proxy models, TRV captures both between‑proxy disagreement and within‑proxy variance, enabling Disagreement‑Regularized Importance Sampling (DR‑IS) and TRV‑based pruning to outperform random selection on noisy CIFAR‑10/100 datasets.
By Csongor Horv\'ath, Ida-Maria Sintorn, Prashant Singh
arXiv:2606. 03650v1 Announce Type: cross Abstract: Choosing or ranking language models for a specific application is hardest when no task-specific labeled data exists, and standard public benchmarks cannot be trusted, their items having likely leaked into pretraining, so scores reflect memorization rather than fitness.
By Alexander Apartsin, Yehudit Aperstein
The paper presents Baszta, a Polish multi‑label content‑safety classifier trained by fine‑tuning the 124M‑parameter allegro/herbert‑base‑cased model on five categories (hate, vulgarity, sexual content, crime, self‑harm) using a Focal + R‑Drop objective. In out‑of‑distribution evaluation on the Gadzi Język benchmark, Baszta achieves a small but statistically significant improvement in micro‑F1 over the Bielik Guard system, though the macro‑F1 advantage disappears when both models are properly tuned. The study also explores calibration techniques, showing that per‑category temperature scaling can recover performance lost by Platt scaling or isotonic regression, and discusses the trade‑offs between robust calibration and adversarial recall.
By Adam G\'orski, Mateusz J\k{a}kalak, Rafa{\l} Jakubowski
The paper presents a cost‑effective approach for industrial explainable‑recommendation systems by decoupling explanation generation from selection. Candidate explanations are pre‑generated using six prompt styles and two commodity LLMs, then a lightweight CPU‑resident selector (e.g., LambdaRank) chooses the best one at request time, achieving sub‑100 ms latency without GPUs. Experiments on a 2,958‑pair Google Local subset and a 300‑pair MovieLens‑1M split show that pairwise ranking methods outperform single‑action RL baselines, while KG‑path selectors achieve near‑perfect user satisfaction scores.
By Tanay Chowdhury, Saeideh Shahrokh Esfahani
arXiv:2609. 18960v1 Announce Type: new Abstract: Quality-aware synthetic-data selection rests on a proxy: examples that an LLM judge rates as good should also help a downstream model learn.
By Son Ha Xuan, Phat T. Tran-Truong, Xuan-Bach Le
The paper presents a cost‑effective approach for industrial explainable‑recommendation systems by decoupling explanation generation from selection. Explanations are pre‑generated using six prompt styles and two commodity LLMs, then a lightweight CPU‑resident selector (e.g., LambdaRank) chooses the best one at request time, achieving sub‑100 ms latency without GPUs. Experiments on a 2,958‑pair Google Local subset and a 300‑pair MovieLens‑1M split show that pairwise ranking outperforms single‑action RL methods, while KG‑path selectors achieve near‑perfect unique‑output rates, and the overall end‑to‑end build cost is around $15 on commodity hardware.
Agentic systems have widened the gap between producing candidate outputs and reviewing them. This paper asks a practical architectural question: should domain specialization be built into an evaluator's weights, or into the rule that decides when its judgment can be trusted?
arXiv:2610.07105v1 Announce Type: cross
Abstract: Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its late...
By Jinfeng Xu, Zheyu Chen, Ziyue Peng, Zheng Lin, Wenhao Yuan, Jian Chen, Shujie Li, Edith Ngai
arXiv:2609.21208v1 Announce Type: new
Abstract: Self-play methods that co-train a single language model as both coder and test author promise to move code-generation RL beyond fixed test suites, but...
By Ana Nunez, Peyman Najafirad
EvoRank is an open autonomous ranking engineer that uses an LLM-guided evolutionary loop to automatically design complete Learning-to-Rank pipelines—including features, models, losses, and ensembles—for multi-objective e-commerce search. On the Expedia ICDM 2013 dataset, EvoRank converged within 50 iterations on interpretable pipelines that outperform an Optuna-tuned LambdaMART on 60k held-out queries and rank in the top 6 % of the original competition. The authors also introduce a headroom gate that predicts whether the evolutionary loop will be worthwhile before any LLM computation, and they release the system, auditing tools, and a catalog of failure modes to help teams apply the method to their own ranking stacks.
By Rayhan Patel, Shabaz Patel
arXiv:2607. 18700v1 Announce Type: cross Abstract: The FathomNetCLEF 2026 competition combines underwater object detection and fine-grained marine species classification under a positive-unlabeled evaluation setting.
By Robert James Brock, Sebastian Maximilian Krupa, Jason Kahei Tam
The study investigates whether Jev, a typed classifier that outputs probabilities over allowed answers without generating text, can replace large language model (LLM) rubric judges. Across nine panels from seven benchmarks, Jev’s accuracy differed significantly from LLM judges in only 8 of 27 paired comparisons, performing best on binary criteria and worse only on graded ones, while most other comparisons were inconclusive. In terms of cost and speed, Jev was 29 to 325 times cheaper and 30 to 220 times faster than the flash‑tier LLM judges, and a cascade approach that defers uncertain Jev verdicts to an LLM yielded only modest gains.
whyItMatters":"The findings suggest that a lightweight classifier like Jev can serve as an efficient first‑stage evaluator, potentially reducing the reliance on expensive and slow LLM judges in automated grading pipelines."
By Delip Rao, Chris Callison-Burch