LoRA, PEFT, instruction tuning and domain adaptation — adapting a pretrained model without paying to train one.
arXiv:2610.27901v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used as research collaborators for hypothesis refinement, methodological planning, and technical analys...
The paper introduces SelfDx, a benchmark that tests whether AI models appropriately respond to health queries by either cautioning against self-diagnosis or agreeing when evidence rules out serious alternatives. It evaluates 27 frontier models on 100 paired cases, finding that most models inappropriately agree in caution-required scenarios, achieving only a 77.0% appropriate caution rate. The study shows that prompt engineering can raise caution but at the cost of agreement, while supervised fine‑tuning on balanced examples improves both metrics.
arXiv:2610.27439v1 Announce Type: cross Abstract: Solving reasoning tasks supported on graphs typically requires providing a Language Reasoning Model (LRM) with a serialized description of the graph....
arXiv:2610.27555v1 Announce Type: cross Abstract: Safety pretraining has emerged as a promising paradigm for embedding safety deeply into a model during pretraining, yielding robustness to subsequent...
The paper shows that in text‑to‑image evaluation, human preference for attractive faces can outweigh objective image quality, leading to biased rankings and model outputs. A user study revealed that annotators consistently rate more attractive faces as higher quality, even when comparing real photographs of equal fidelity. This bias can invert leaderboard rankings, dominate preference model scores, and cause generated faces to shift toward younger, more feminine features, undermining user control.
The paper introduces CETBench, a benchmark for assessing large language models (LLMs) on code-equivalence checking by generating program pairs through random code transformations. It demonstrates that even simple transformations significantly degrade LLM performance, especially in cross-lingual scenarios, and proposes a fine-tuning approach to improve results. The study provides detailed ablations on program difficulty and transformation types, offering insights into LLMs’ limited semantic understanding of code.
The paper introduces Adaptive Dual-Teacher Distillation with Subnetwork Rectification (DDSR), a method for black-box domain adaptation that fuses predictions from a black-box source model and a vision‑language model to generate reliable pseudo‑labels for the target domain. DDSR uses adaptive prediction fusion, subnetwork regularization to prevent overfitting, iterative refinement of pseudo‑labels and prompts, and class‑wise prototypes for self‑training. Experiments on several benchmark datasets show that DDSR outperforms existing state‑of‑the‑art approaches, even those that have access to source data or source model parameters.
The paper introduces CultureLens, a benchmark with 9,261 scenarios from 75 countries that tests how large language models (LLMs) incorporate cultural information. Experiments on Llama‑3.1‑8B, Gemma‑3‑12B, and Qwen‑3.5‑27B show that presenting a contradictory cultural norm reduces aligned answers by 32‑47 percentage points, even when the model knows the correct country. Fine‑tuning with contradictory‑norm examples improves alignment to 85‑93% under contradictory norms while only slightly affecting performance under true norms.
The paper introduces SHAM-Bench, a unified benchmark for detecting span-level hallucinations across six natural language generation tasks, compiled from 6,461 human-annotated examples under a common schema. It evaluates existing detection methods, finding that fine‑tuned detectors perform well in‑domain but falter under distribution shift, while prompt‑based methods with reasoning show greater robustness. The study also highlights challenges such as subtle unsupported content, ambiguous span boundaries, and noisy labels, and releases the dataset, evaluation code, and a public leaderboard for future research.
Self-Decomposing Diffusion for Code (SDDC) is a new diffusion language model framework that first generates a contract outlining program decomposition and inter‑unit dependencies, then uses this contract to guide coordinated parallel implementation of complex programs. By planning decomposition before generation, SDDC balances the trade‑off between aggressive parallel decoding and maintaining program correctness. Across five benchmarks and two 7B diffusion backbones, SDDC improves pass@1 by up to 31.1 percentage points while decoding 7.5–13.7× faster than base models, and outperforms a fine‑tuning baseline by up to 23.5 points with lower latency.
The study investigates how progressive neuronal loss affects language models by systematically ablating computational units in LSTM and Transformer architectures. LSTMs show a gradual decline in syntactic and semantic abilities, while Transformers exhibit a threshold-like collapse under MLP ablation and a smoother sigmoid-like decline under attention-head ablation. These distinct failure modes mirror different aphasic syndromes, suggesting that architecture-dependent degradation patterns can provide computational insights into neurodegenerative diseases.
The paper introduces RoboJudge, a multimodal language model trained to evaluate embodied video generation. It presents RoboJudge-Bench, a benchmark of 800 human‑annotated videos assessing Physical Adherence and Instruction Alignment, and RoboJudge-Train, a 12K‑video dataset with human‑verified scores and rationales. Fine‑tuned on this data, RoboJudge‑9B achieves the highest agreement with human judgments, outperforming existing models and improving performance on external physics benchmarks.
DriveVLA-M0 introduces a failure-aware latent memory system for Vision‑Language‑Action models in autonomous driving. It stores past failure cases with scene representations and expert trajectories, and uses a retrieval model to decouple static road structure from dynamic agent interactions. During inference, retrieved cases are injected via a lightweight LoRA‑based test‑time training mechanism, yielding significant performance gains on NAVSIM benchmarks with minimal latency overhead.
S$^2$-MAS is a society‑scale multi‑agent simulation framework that models 500 million synthetic household personas based on real‑survey data. Its Genese engine separates decision making from state updates, while FinL‑mlx, a compact decision model trained on 2.93 million examples, achieves high accuracy in consumer financial‑choice scenarios. The framework evaluates eight financial task families and simulates 12 million distinct agents, revealing how purchase goals, credit limits, and demographic factors influence consumption, cash flow, debt repayment, and arrears over a 24‑month panel.
arXiv:2608.05387v1 Announce Type: cross Abstract: Language models can prepare VASP INCAR files from natural-language requests, but so far only large proprietary cloud models come close to handling th...
arXiv:2610.27490v1 Announce Type: cross Abstract: Mitigating catastrophic risks from large language models requires excising hazardous operational capabilities, such as cyberwarfare exploitation and...
arXiv:2610.27698v1 Announce Type: cross Abstract: Vision-language-action (VLA) models transfer knowledge from large-scale pretraining into robot control, but their manipulation performance remains se...
arXiv:2610.28407v1 Announce Type: cross Abstract: We present our solution to the AKBC LM-KBC 2026 Shared Task, where we achieved a score of 0.6836 on the official leaderboard, more than double the ba...
arXiv:2610.28622v1 Announce Type: cross Abstract: Spatial intelligence has recently been assessed through generative tasks such as spatially grounded image editing, yet language-guided generation of...
arXiv:2512.23067v3 Announce Type: replace Abstract: Personalized alignment from preference data has focused primarily on improving personal reward model (RM) accuracy, with the implicit assumption th...