EmoDistill is an offline framework that distills emotional negotiation skills from large language model interactions into smaller agents. It separates emotion selection, handled by an Implicit Q‑Learning selector, from emotion‑conditioned expression, learned by a LoRA‑adapted 7B policy via supervised fine‑tuning and judge policy optimization. Experiments across four negotiation domains show that the full EmoDistill policy outperforms vanilla and IQL‑only baselines, while removing the explicit emotion channel markedly reduces negotiation utility and reveals partial, domain‑dependent transfer to unseen counterparties.
By Yunbo Long, Haolang Zhao, Lukas Beckenbauer, Liming Xu, Alexandra Brintrup
The paper investigates using synthetic natural-language descriptions to contrastively pretrain small transformer encoders for code representation. By pairing generated descriptions with code in a dual-encoder setup during training and discarding them at inference, the authors achieve significant improvements over traditional pretraining baselines on most evaluated tasks. When fine‑tuned, these models match or surpass much larger zero‑shot models and remain competitive with execution‑aware supervision, indicating a scalable alternative for code embeddings.
By Kenneth Paulsen, Florian Tambon, Mike Papadakis, Shin Yoo
IDSPACE is a new synthetic document generator designed to aid the evaluation of digital identity verification systems. It uses model‑guided Bayesian optimization to tune generation parameters for visual similarity and prediction consistency, separates user‑defined metadata from low‑level control settings, and supports scanned and mobile‑captured documents. Experiments demonstrate that IDSPACE improves evaluation consistency by 15‑45 % over baselines, boosts training accuracy by up to 9 %, and increases SSIM similarity by 10 %, while providing a dataset of 359,240 synthetic documents across ten European ID types.
By Lulu Xie, Yancheng Wang, Kanchan Chowdhury, Rolando Garcia, Yingzhen Yang, Jia Zou
The paper investigates weight‑space merging of independently fine‑tuned multilingual machine translation models. Experiments show that merging is more successful when models share a target language, yet it still cannot match the peak performance of language‑specific checkpoints. When target languages differ, performance drops sharply, and analysis reveals that overlapping neuron activation and incompatible upper‑layer geometries cause these failures.
By Baban Gain, Trilok Nath Singh, Asif Ekbal
The paper presents an end‑to‑end sequence‑to‑sequence approach for correcting Tamil spelling and grammar errors, leveraging progressively fine‑tuned transformer models (mT5‑small and mBART‑50). Using a synthetic corpus of 657,720 noisy‑clean sentence pairs across ten error categories, the authors introduce a four‑stage training schedule that targets surface noise, contextual grammar, single‑site sandhi, and multi‑site cross‑word sandhi. The best model, mBART‑50 v5, achieves 69.3% exact‑match accuracy on a balanced diagnostic set, with notable gains in sandhi (87.5%) and subject‑verb agreement (43.5%) accuracy, while also revealing a precision‑recall trade‑off for sandhi corrections.
By Karthikeyan A, Jaya Nirmala S, Sangeetha Sivanesan, Indhu R, Pranav Kumar, Bharat Jude Johnson, Vishnu Ram
LightEMMA is a longitudinal evaluation framework that tests the autonomous driving performance of vision‑language models (VLMs) without fine‑tuning or prompt engineering. Using this protocol, the authors evaluated 15 VLMs from five major families on the nuScenes prediction benchmark and found that larger, more capable models do not consistently outperform earlier generations. The study identifies common failure modes such as overreliance on historical actions and difficulty reconciling conflicting visual cues, underscoring the need for domain‑specific adaptation to enhance VLM safety in autonomous driving.
By Zhijie Qiao, Haowei Li, Zhong Cao, Henry X. Liu
Terminal-Universe is a framework that transforms large collections of terminal‑based agent trajectories into reusable, executable environments. By replaying recorded file operations and completing missing files, it reconstructs the original workspace and task, then synthesizes new tasks and multi‑round interactions. The resulting 37.3k task‑sufficient environments enable significant performance gains when fine‑tuning language models on terminal‑based benchmarks.
By Jie Wu, Zhenru Zhang, Beichen Zhang, Xuwu Wang, Yuhui Su, Mouxiang Chen, Peng Wang, Zhihai Wang, Que Shen, Hao Zhou, An Yang, Fei Huang, Yujiu Yang, Dayiheng Liu
Jina-OCR-v1 is an end‑to‑end document parsing model designed for low‑budget GPUs, combining a compressed‑vision encoder with a 3B mixture‑of‑experts decoder that activates about 570 M parameters per token. It uses a FastMTP speculative decoding head that shares a single draft block across three prediction steps, with greedy verification ensuring lossless decoding. Post‑training includes instruction alignment, robustness fine‑tuning on difficult documents, and GRPO with dense verifiable rewards, achieving 91.14 on OmniDocBench v1.6 and 83.4 on olmOCR‑Bench while delivering the highest page throughput at 2.57 pages per second on an NVIDIA L4 GPU.
By Alejandro Bar\'on Garc\'ia, Feng Wang, Emilia Garcia Casademont, Han Xiao
AgentRM proposes a generalizable reward model to guide LLM-based agents during test-time search, outperforming direct policy fine-tuning. Three reward modeling strategies—explicit, implicit, and LLM-as-a-judge—are explored, and AgentRM improves base policy performance by an average of 8.8 points across nine tasks, surpassing top general agents by 4.0 points. It also shows strong weak-to-strong generalization and can boost specialized agents, with plans to release code for further research.
By Yu Xia, Jingru Fan, Weize Chen, Siyu Yan, Xin Cong, Zhong Zhang, Yaxi Lu, Yankai Lin, Zhiyuan Liu, Maosong Sun
TokenMatch is a transformer-based model that estimates 3D shape correspondences by adaptively tokenising meshes into curvature-guided patches. Trained only on the BeCoS partial-to-partial dataset, it generalises to full-shape matching without retraining, using self‑ and cross‑attention to learn patch‑ and point‑level relations. Evaluated on CP2P, PSMAL, BeCoS, FAUST, SCAPE, and SHREC'19, TokenMatch consistently outperforms existing methods in mean geodesic error and intersection‑over‑union while achieving sub‑second inference speeds.
By Adeela Islam, Zorah L\"ahner, Vittorio Murino, Vladislav Golyanik
SafeRI proposes an on-demand safety alignment approach for large vision-language models, contrasting with existing always-on methods that globally modify model behavior. The framework uses a lightweight recognizer to evaluate token-level safety during autoregressive generation, gating a LoRA module that only activates when unsafe content is detected. By training the LoRA on unsafe prefixes and safe continuations, SafeRI redirects unsafe generations back to safe responses without perturbing the model’s original reasoning path.
By Caoyuan Ma, Tian Gu, Wenpu Liu, Weichu Xie, Shuai Dong, Yuqi Xu, Ji Zhao, Ziyue Wang, Wenzheng Chang, Taiqiang Wu, Yongfu Zhu, Wenqi Shao, Zheng Wang, Yinqiang Zheng
The paper introduces a benchmark and evaluation system for measuring how well generative image models preserve the identity of a subject across generation, editing, restoration, and multi‑subject scenarios. It compares three paradigms—input context, trainable subject‑specific parameters, and a persistent identity layer—showing that persistent identity consistently improves fidelity while keeping image quality and instruction adherence high. The study finds that identity preservation remains a distinct limitation of current foundation models, especially under iterative edits, small scales, severe degradation, and multi‑subject composition.
By Mengwei Ren, Xuaner Zhang, Zhihao Xia
GraFT is a training‑free framework that enhances spatial reasoning in multimodal large language models by integrating a compact 3D scene graph (3DSG). It offers deterministic geometry via symbolic tools, allocentric layout through bird’s‑eye‑view rendering, and visual‑attribute grounding using egocentric frames. Experiments on ScanQA and VSI‑Bench show significant performance gains, with CIDEr increasing by 27% and improvements up to 65% over baseline models.
By Junqing Du, Fernando Ropero, Erkin Turkoz, Yanfeng Zhang, Lu Liu
The paper investigates how large language models learn new tasks in-context, comparing rule-based instruction following to example-based few-shot prompting across five diverse tasks. Results show that models generally learn more reliably from rule descriptions than from examples alone, and adding more examples does not consistently improve performance. Instruction tuning further enhances rule-based learning while preserving example-based capabilities, with rule advantages being strongest for algebraic tasks and weaker for tasks requiring distributional sensitivity or parametric knowledge.
By Xiang Fu, Seungmin Cho, Yukyung Lee, Najoung Kim
The paper introduces Linearized Subspace Refinement (LSR), a post‑training framework that uses the local linearized model of a trained neural network to compute a low‑dimensional correction via a reduced least‑squares problem. LSR is architecture‑agnostic and improves accuracy across tasks such as function approximation, operator learning, physics‑informed fine‑tuning, and noisy inverse problems, often achieving order‑of‑magnitude error reductions. The method reveals that standard training can leave significant accuracy plateaus due to numerical ill‑conditioning, and it offers a subspace rank that balances correction strength, stability, and noise sensitivity.
By Wenbo Cao, Weiwei Zhang
The paper investigates how on-policy distillation (OPD) and reinforcement learning with verifiable rewards (RLVR) can be combined for post‑training reasoning in large language models. It shows that a two‑stage approach—first applying OPD, then RL—outperforms single‑signal methods and other joint baselines on logic and math reasoning benchmarks. The authors explain this advantage through pass@k analysis, learning dynamics, and parameter updates, concluding that OPD expands solution coverage while RL sharpens performance within that support, and that the OPD validation score is the key trigger for switching to RL.
By Boyan Li, Bingsen Chen, Chenghao Yang, Ping Nie, Chen Zhao, Xi Ye
The paper introduces LDC, a framework that learns to generate research ideas with dynamic control. It combines supervised fine‑tuning on paper‑idea pairs with controllable reinforcement learning that optimizes novelty, feasibility, and effectiveness. During inference, sentence‑level controllers steer the generation process to balance these dimensions.
By Ruochen Li, Liqiang Jing, Chi Han, Jiawei Zhou, Xinya Du
The paper investigates how the phrasing of prompts affects large language models (LLMs) in predicting drug toxicity. By varying job role, prompt structure, and rule interpretation, the authors found that natural variability in LLM outputs outweighs fine‑tuning of prompts. However, incorporating chemoinformatic code to extract features significantly improved model performance, suggesting that prompt engineering alone is insufficient for reliable toxicity prediction.
By Mia MacGregor, Aakash Welgamage Don, Mark Bartlett
Para‑Pipe is a hierarchical mapping framework that combines intra‑ and inter‑stage operator parallelism within a pipelined architecture to optimize deep‑learning inference on heterogeneous System‑on‑Chip (SoC) platforms. By selectively tuning parallelism levels across pipeline stages, it balances throughput and latency while reducing inter‑processor communication overhead. Evaluations on Amlogic and Black Sesame SoCs show Pareto‑optimal configurations, with throughput‑optimized settings achieving up to 11.0 % higher energy efficiency than purely pipelined approaches and 23.3 % over non‑pipelined parallel execution.
Terminal-Universe is a framework that converts large collections of terminal-based agent trajectories into reusable, executable environments. By replaying recorded file operations and filling missing files with a completion agent, it reconstructs the original workspace and generates new tasks, scaling them in breadth (cross-workspace queries) and depth (multi-round interactions). The resulting 37.3k task-sufficient environments enable significant performance gains when fine‑tuning language models on terminal‑centric benchmarks.