Direct-copy and negative-control analyses show that gains depend on source-target compatibility and LLM adaptation; stored source accuracy alone does not predict transferability.
Abstract
Neural-network repositories contain executable models, recipes, input transformations, and measured accuracies. We study whether one same-family experiment can be curated as prompt guidance for LLM-based improvement of a low-performing target under equal generation and evaluation budgets. TuneNNGen extends NNGPT with a source-guided route and compares it with target-only generation on one CIFAR-10 target, two fixed source-selection rules, three code LLMs, and an additional SVHN target. Under the historical one-epoch search protocol, best-of-budget accuracy on the available evaluation split rises from 23.98% to 50.49% on CIFAR-10 and from 22.54% to 78.80% on SVHN. Selected five-epoch, three-seed means on train-derived validation splits retain gains of 40.94 points on CIFAR-10, 18.83 on Imagenette, and 7.27 on CIFAR-100. Direct-copy and negative-control analyses show that gains depend on source-target compatibility and LLM adaptation; stored source accuracy alone does not predict transferability.
Competitive programming has become a key test of large language model reasoning, with international competitions such as IOI and ICPC representing its most challenging settings. We present an end-to-end specialization pipeline combining large-scale problem curation, synthetic reasoning traces, supervised fine-tuning (SFT), and reinforcement learning (RL). Using 22,000 curated problems, we train Nemotron-3-Nano-CC (30B-A3B) with SFT and RL and Nemotron-3-Ultra-CC (550B-A55B) with SFT alone. We further introduce GenCorrect, a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. On IOI 2025, Nano-CC improves from 130 points to 291 after post-training and to 468 with GenCorrect, exceeding the gold threshold of 438.3 while Ultra-CC reaches 502. Guided by these results, we develop a competition-specific Ultra-CC system and evaluate it prospectively during IOI 2026. Under the same time, internet-access, and submission constraints as human contestants, it scores 535.4 out of 600, exceeding both the gold threshold of 361.12 and the top human score of 498.27. To our knowledge, this is the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
Aleksander Ficek, Sean Narenthiran, Mehrzad Samadi et al.· 0 citations
This paper evaluates various machine learning approaches on Eclipse dataset including using transformer based architectures with frozen weights as embedding with a fine-tuned neural network classification head (MLP) and experimented with Retrieval Augmented Generation (RAG).
Márk Lajkó, Balázs Nagy, László Vidács· International Conference on...· 0 citations
Test Case Prioritization (TCP) aims to reorder test cases to enable earlier fault detection in Continuous Integration pipelines. Machine learning-based TCP approaches have shown promising results but require substantial historical execution data, creating a cold-start challenge for new or rapidly evolving projects. This paper presents an empirical study of direct transfer learning, where models trained on data-rich source projects are applied to unseen targets without any retraining. A total of 578 experiments are conducted across 17 Defects4J projects (34 within-project baselines and 544 cross-project transfer scenarios) using 270,611 feature vectors. Results show that with appropriate source selection, cross-project transfer can match or surpass within-project baselines. A three-way feature ablation (HIS-only, COV-only, full 13-feature) finds that no feature subset consistently exceeds random ordering (E[APFD]≈0.5) on average, confirming that source selection is the critical variable. Analysis identifies failure-rate alignment and source dataset richness as key predictors of transfer quality, providing source-selection criteria that require no target-side data collection.
Michael Johnson, Khaled Slhoub· International Conference on...· 0 citations
The engineering details of ADEPT are presented, a framework that integrates representative adequacy techniques, including neuron-coverage-based metrics, surprise adequacy, input distribution coverage, boundary coverage, and source- and model-level mutation score, under a consistent execution workflow.
Yidi Kao, Shawn Burnham, Tommi Rose Fahy et al.· 0 citations
Treating supervision format as a first-class hyperparameter for multi-task reasoning SFT in large language models—at least in this benchmark-and-model setting—rather than a mere rendering detail is supported.
Nhat Thanh Vu, M. Rashid, Fariza Sabrina· Electronics· 0 citations
FlavourBench is introduced, which instead compiles dense answer maps from a versioned culinary environment and remains similar under alternative metrics, task filters, family weights, and three public Epicure checkpoints.
J. Chen, Erim Hayretci· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
