{"total":10,"items":[{"citing_arxiv_id":"2605.15044","ref_index":35,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"SpeakerLLM: A Speaker-Specialized Audio-LLM for Speaker Understanding and Verification Reasoning","primary_cat":"cs.SD","submitted_at":"2026-05-14T16:36:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SpeakerLLM unifies speaker profiling, recording-condition understanding, and structured verification reasoning in an audio-LLM via a hierarchical tokenizer and decision traces.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13711","ref_index":74,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"MILM: Large Language Models for Multimodal Irregular Time Series with Informative Sampling","primary_cat":"cs.LG","submitted_at":"2026-05-13T15:58:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MILM fine-tunes LLMs on XML-encoded multimodal irregular time series via a two-stage process that exploits informative sampling patterns to achieve top performance on EHR classification datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12368","ref_index":26,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"MetaColloc: Optimization-Free PDE Solving via Meta-Learned Basis Functions","primary_cat":"cs.LG","submitted_at":"2026-05-12T16:36:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MetaColloc meta-learns a universal set of neural basis functions offline so that new PDEs can be solved at test time with a single linear solve instead of per-equation neural-network optimization.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"Algorithm 1Meta-Training the Neural Basis Dictionary Require:Meta-training epochsE, tasks per epochT, neural networkΦ θ 1:forepoch= 1toEdo 2:fortask= 1toTdo 3:Sample inputsXand targetsYfrom multi-scale GRF distribution 4:Compute basis matrixΦ θ(X) 5:Solve for coefficients:w=lstsq(Φ θ(X), Y) 6:Predict outputs: ˆY= Φ θ(X)w 7:Compute MSE loss between ˆYandY 8:Updateθwith AdamW [26] 9:end for 10:end for 11:returnFrozen networkΦ frozen 3.4 optimization-free Collocation for Linear PDEs After meta-training, the network Φfrozen is ready for PDE solving. We use it inside a collocation framework. Consider the Poisson equation −∆u=f with boundary condition u=g . We sample N interior points and Nb boundary points. We pass these points through the frozen network to get basis"},{"citing_arxiv_id":"2605.09439","ref_index":24,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Inverse Design for Conditional Distribution Matching","primary_cat":"cs.LG","submitted_at":"2026-05-10T09:27:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Defines Conditional Distribution Matching (CDM) as finding inputs whose induced conditional distributions match a target distribution and proposes the MLGD-F inference-time algorithm using pretrained diffusion models to solve it without retraining.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Information Processing Systems, volume 30, pages 2203-2213, 2017. [22] Yujia Li, Kevin Swersky, and Richard Zemel. Generative moment matching networks. In International Conference on Machine Learning, volume 37, pages 1718-1727. PMLR, 2015. [23] Shanchuan Lin, Anran Wang, and Xiao Yang. Sdxl-lightning: Progressive adversarial diffusion distillation, 2024. URLhttps://arxiv.org/abs/2402.13929. [24] Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. InInternational Conference on Learning Representations, 2019. URL https://openreview.net/forum? id=Bkg6RiCqY7. [25] Colin McDiarmid. On the method of bounded differences. In J. Siemons, editor,Surveys in Combinatorics, 1989, volume 141 ofLondon Mathematical Society Lecture Note Series, pages"},{"citing_arxiv_id":"2605.09269","ref_index":26,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"DeltaRubric: Generative Multimodal Reward Modeling via Joint Planning and Verification","primary_cat":"cs.CL","submitted_at":"2026-05-10T02:32:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DeltaRubric decomposes multimodal preference evaluation into self-generated planning and verification steps within a single model, producing large accuracy improvements on VL-RewardBench via multi-role reinforcement learning.","context_count":1,"top_context_role":"other","top_context_polarity":"unclear","context_text":"Song, Haitao Mi, Pratap Tokekar, et al. V ogue: Guiding exploration with visual uncertainty improves multimodal reasoning.arXiv preprint arXiv:2510.01444, 2025. [25] Tianci Liu, Ran Xu, Tony Yu, Ilgee Hong, Carl Yang, Tuo Zhao, and Haoyu Wang. Openrubrics: Towards scalable synthetic rubric generation for reward modeling and llm alignment.arXiv preprint arXiv:2510.07743, 2025. [26] Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. InInternational Conference on Learning Representations, 2019. URL https://openreview.net/forum? id=Bkg6RiCqY7. [27] Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to"},{"citing_arxiv_id":"2605.07315","ref_index":26,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"LaTER: Efficient Test-Time Reasoning via Latent Exploration and Explicit Verification","primary_cat":"cs.CL","submitted_at":"2026-05-08T06:23:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"LaTER reduces LLM token usage 16-33% on reasoning benchmarks by exploring in latent space then switching to explicit CoT verification, with gains like 70% to 73.3% on AIME 2025 in the training-free version.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"and stronger on most other tasks. This indicates that supervised latent training improves the overall accuracy-efficiency frontier but does not uniformly dominate every benchmark. We view this as evidence that latent-budget allocation and data mixture remain important design choices. 4 Related Works Training-free latent reasoning.Soft Thinking and SwiReasoning [ 26, 27] replace hard token inputs with probability-weighted mixtures of token embeddings, enabling latent reasoning from the model's own next-token distribution. However, soft-embedding methods can collapse toward the dominant token and thus behave similarly to greedy decoding, limiting their ability to maintain alternative reasoning paths. SeLaR [ 28] addresses this issue with entropy-gated activation, applying latent"},{"citing_arxiv_id":"2605.07276","ref_index":23,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Signal Reshaping for GRPO in Weak-Feedback Agentic Code Repair","primary_cat":"cs.AI","submitted_at":"2026-05-08T05:41:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Reshaping outcome rewards, process signals, and rollout comparability in GRPO raises strict compile-and-semantic accuracy in agentic code repair from 0.385 to 0.535 under weak feedback.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Conference on Learning Representations, 2019. URL https://openreview.net/forum? id=Bkg6RiCqY7. [22] Liangchen Luo, Yinxiao Liu, Rosanne Liu, Samrat Phatale, Meiqi Guo, Harsh Lara, Yunxuan Li, Lei Shu, Yun Zhu, Lei Meng, Jiao Sun, and Abhinav Rastogi. Improve mathematical reasoning in language models by automated process supervision.arXiv preprint arXiv:2406.06592, 2024. [23] Michael Luo, Sijun Tan, Colin Cai, Roy Hao, Lambda Lu, Tarun Venkat, Tianjun Zhang, Manan Roongta, Tianhao Wu, Justin Wong, Ion Stoica, and Raluca Ada Popa. DeepSWE: Training a fully open-sourced, state-of-the-art coding agent by scaling RL. Together AI Blog, 2025. URL https://www.together.ai/blog/deepswe. Agentica Project, UC Berkeley Sky Computing Lab and Together AI."},{"citing_arxiv_id":"2605.06158","ref_index":18,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Stateful Agent Backdoor","primary_cat":"cs.CR","submitted_at":"2026-05-07T12:48:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A stateful backdoor for LLM agents, modeled as a Mealy machine with a decomposition framework, enables incremental malicious actions across sessions and achieves 80-95% attack success rate on four models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05652","ref_index":52,"ref_count":2,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Information-Preserving Domain Transfer with Unlabeled Data in Misspecified Simulation-Based Inference","primary_cat":"cs.LG","submitted_at":"2026-05-07T04:06:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SPIN performs bidirectional domain transfer in SBI to retain parameter mutual information from unlabeled real observations, improving real-world posterior inference under increasing misspecification.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.21106","ref_index":26,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"How Much Is One Recurrence Worth? Iso-Depth Scaling Laws for Looped Language Models","primary_cat":"cs.LG","submitted_at":"2026-04-22T21:51:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A fitted iso-depth scaling law measures that one recurrence in looped transformers is worth r^0.46 unique blocks in validation loss.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}