{"total":32,"items":[{"citing_arxiv_id":"2605.19048","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Conserved Kinematic Representations enable Zero-Shot Decoding in Handwriting BCIs","primary_cat":"q-bio.NC","submitted_at":"2026-05-18T19:13:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A zero-shot machine learning decoder for handwriting BCIs achieves 64% hits@3 retrieval on unseen letters by exploiting conserved kinematic neural representations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18507","ref_index":1,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Weakly Supervised Cross-Modal Learning for 4D Radar Scene Flow Estimation","primary_cat":"cs.CV","submitted_at":"2026-05-18T15:00:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A task-specific iterative framework for weakly supervised 4D radar scene flow estimation uses instance-aware self-supervised losses from 2D tracking/segmentation and a rigid static loss from odometry to outperform LiDAR-dependent cross-modal and fully supervised methods on the VoD dataset.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17187","ref_index":257,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media","primary_cat":"cs.CL","submitted_at":"2026-05-16T22:52:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"PluRule is a new multimodal multilingual benchmark showing that state-of-the-art vision-language models perform only marginally better than a trivial baseline at detecting specific rule violations in pluralistic online communities.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Lampe, Jacob Eisenstein, and Eric Gilbert. 2018. The internet's hidden rules: An empirical study of reddit norm violations at micro, meso, and macro scales. Proceedings of the ACM on Human-Computer Inter- action, 2(CSCW):1-25. Joseph Paul Cohen and Henry Z. Lo. 2014. Aca- demic Torrents: A Community-Maintained Dis- tributed Repository. InProceedings of the 2014 Annual Conference on Extreme Science and Engi- neering Discovery Environment, XSEDE '14, pages 1-2, New York, NY , USA. Association for Comput- ing Machinery. Ángel Díaz and Laura Hecht-Felella. 2021. Double standards in social media content moderation. Tech- nical report, Brennan Center for Justice at New York University School of Law. Accessed: 24 Dec 2025."},{"citing_arxiv_id":"2605.12335","ref_index":23,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"EHR-RAGp: Retrieval-Augmented Prototype-Guided Foundation Model for Electronic Health Records","primary_cat":"cs.IR","submitted_at":"2026-05-12T16:17:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"EHR-RAGp is a retrieval-augmented EHR foundation model that employs prototype-guided retrieval to dynamically integrate relevant historical patient context, outperforming prior models on clinical prediction tasks.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Traditional EHR modeling approaches process patient trajectories as sequences of discrete clinical events collected over time. Some models primarily consider the occurrence of certain events, while others process the occurrence of the clinical event and its measured value. Early techniques utilized recurrent models such as Long-Short Term Memory (LSTM) [22] or Gated Recurrent Unit (GRU) networks [23] to capture long-term dependencies [24, 25, 16, 26]. 2 The emergence of transformer-based LLMs [27, 28] paved the way for more advanced EHR foun- dation models. For example, encoder-based models such as BEHRT [29], Hi-BEHRT [30], CEHR- BERT [31], and Med-BERT [32] introduced large-scale pretraining over longitudinal sequences of medical events to enhance the contextualization of patient trajectories across visits."},{"citing_arxiv_id":"2605.03045","ref_index":270,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TCD-Arena: Assessing Robustness of Time Series Causal Discovery Methods Against Assumption Violations","primary_cat":"cs.LG","submitted_at":"2026-05-04T18:12:33+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"TCD-Arena is a new customizable testing framework that runs millions of experiments to map how 33 different assumption violations affect time series causal discovery methods and shows ensembles can boost overall robustness.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.00553","ref_index":43,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance","primary_cat":"cs.LG","submitted_at":"2026-05-01T10:42:08+00:00","verdict":null,"verdict_confidence":null,"novelty_score":null,"formal_verification":null,"one_line_summary":null,"context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.14073","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Neural architectures for resolving references in program code","primary_cat":"cs.LG","submitted_at":"2026-04-15T16:44:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"New seq2seq architectures for permutation indexing outperform baselines on synthetic reference-resolution tasks and reduce real decompilation error rates by 42%.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.12825","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"The illusory simplicity of the feedforward pass: evidence for the dynamical nature of stimulus encoding along the primate ventral stream","primary_cat":"q-bio.NC","submitted_at":"2026-04-14T14:48:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Primate ventral stream encodes visual stimuli through evolving neural dynamics that carry category information beyond any fixed spatial pattern during the initial feedforward pass.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.07090","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Leveraging Artist Catalogs for Cold-Start Music Recommendation","primary_cat":"cs.IR","submitted_at":"2026-04-08T13:43:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ACARec attends over artist catalogs to generate CF embeddings for new tracks, more than doubling recall and NDCG versus content-only baselines in music recommendation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.06150","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Delta6: A Low-Cost, 6-DOF Force-Sensing Flexible End-Effector","primary_cat":"cs.RO","submitted_at":"2026-04-07T17:52:32+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Delta6 delivers a low-cost 6-DOF force-sensing end-effector with 3.8% FS accuracy using sequence models, validated on robot-arm tasks like buffing and tight assembly.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.14360","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"M$^2$RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling","primary_cat":"cs.LG","submitted_at":"2026-03-15T12:53:09+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"M²RNN achieves perfect state tracking at unseen lengths and outperforms Gated DeltaNet hybrids by 0.4-0.5 perplexity on 7B models with 3x smaller recurrent states.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.05416","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Reduced-Order Surrogates for Forced Flexible Mesh Coastal-Ocean Models","primary_cat":"cs.CE","submitted_at":"2026-02-05T07:59:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Koopman autoencoders with forcings and temporal unrolling deliver accurate year-long predictions for coastal-ocean models at 300-1400x speedup, outperforming POD in two of three cases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2601.23154","ref_index":46,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"On Safer Reinforcement Learning for Sedation and Analgesia in Intensive Care","primary_cat":"cs.LG","submitted_at":"2026-01-30T16:40:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Offline RL for ICU sedation shows that adding 30-day mortality to the objective yields policies whose clinician agreement correlates negatively with mortality, unlike pain-only versions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2601.14925","ref_index":10,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Fast-ULCNet: A fast and ultra low complexity network for single-channel speech enhancement","primary_cat":"eess.AS","submitted_at":"2026-01-21T12:14:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Fast-ULCNet matches original ULCNet speech enhancement quality while cutting model size by more than half and latency by 34% via FastGRNN replacement and a state-drift filter.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2512.13684","ref_index":19,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Recurrent Video Masked Autoencoders","primary_cat":"cs.CV","submitted_at":"2025-12-15T18:59:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"RVM uses recurrent computation inside a masked autoencoder to learn video representations that match or exceed prior video and image models on classification, tracking, and dense spatial tasks with up to 30x better parameter efficiency.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2509.22630","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"StateX: Enhancing RNN Recall via Post-training State Expansion","primary_cat":"cs.CL","submitted_at":"2025-09-26T17:55:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"StateX post-trains RNNs to expand recurrent state size, improving recall and in-context learning with negligible parameter growth.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2508.08441","ref_index":76,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"SpectraLLM: Uncovering the Ability of LLMs for Molecular Structure Elucidation from Multi-Spectral Data","primary_cat":"q-bio.QM","submitted_at":"2025-08-04T13:33:38+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SpectraLLM is an LLM fine-tuned to predict small-molecule structures from single or multiple spectra, reporting state-of-the-art results on four public benchmarks with gains from multi-modal input.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.12590","ref_index":12,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"From Time-series Generation, Model Selection to Transfer Learning: A Comparative Review of Pixel-wise Approaches for Large-scale Crop Mapping","primary_cat":"cs.CV","submitted_at":"2025-07-16T19:18:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A comparative review with experiments identifying optimal preprocessing, models, and transfer strategies for large-scale pixel-wise crop mapping using Landsat 8 data across five sites.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2402.06196","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Large Language Models: A Survey","primary_cat":"cs.CL","submitted_at":"2024-02-09T05:37:09+00:00","verdict":"ACCEPT","verdict_confidence":"MODERATE","novelty_score":3.0,"formal_verification":"none","one_line_summary":"The paper surveys key large language models, their training methods, datasets, evaluation benchmarks, and future research directions in the field.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"The embedding vectors learned by NLMs define a hidden space where the semantic similarity between vectors can be readily computed as their distance. This opens the door to computing semantic similarity of any two inputs regardless their forms (e.g., queries vs. documents in Web search [17], [18], sentences in different languages in machine translation [19], [20]) or modalities (e.g., image and text in image captioning [21], [22]). Early NLMs are task-specific models, in that they are trained on task-specific data and their learned hidden space is task-specific. Pre-trained language models (PLMs), unlike early NLMs, are task-agnostic. This generality also extends to the learned arXiv:2402.06196v3 [cs.CL] 23 Mar 2025"},{"citing_arxiv_id":"2308.08089","ref_index":154,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory","primary_cat":"cs.CV","submitted_at":"2023-08-16T01:43:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DragNUWA integrates text, image, and trajectory controls into a diffusion video model using a Trajectory Sampler, Multiscale Fusion, and Adaptive Training to enable fine-grained open-domain video generation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2212.08989","ref_index":225,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Deep learning applied to computational mechanics: A comprehensive review, state of the art, and the classics","primary_cat":"cs.LG","submitted_at":"2022-12-18T02:03:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A comprehensive review of deep learning techniques for computational mechanics, including LSTM for constitutive modeling, PINNs for PDE solving, optimizers, and kernel methods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2109.10616","ref_index":7,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Enriching and Controlling Global Semantics for Text Summarization","primary_cat":"cs.CL","submitted_at":"2021-09-22T09:31:50+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A normalizing-flow neural topic model plus control mechanism are added to Transformer summarizers to supply and regulate global semantics, with reported gains over prior models on five benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1907.08440","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Neural Cross-Domain Collaborative Filtering with Shared Entities","primary_cat":"cs.IR","submitted_at":"2019-07-19T10:04:28+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"NeuCDCF is a wide-and-deep neural architecture for cross-domain collaborative filtering that jointly learns matrix factorization and deep representations, reporting better performance than prior CDCF models on four real-world datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1907.03609","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Variational Context: Exploiting Visual and Textual Context for Grounding Referring Expressions","primary_cat":"cs.CV","submitted_at":"2019-07-08T13:37:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A variational Bayesian framework exploits reciprocity between referents and context plus semantic reproduction to improve referring expression grounding over pairwise methods in supervised and unsupervised settings.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1907.02050","ref_index":37,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Reasoning and Generalization in RL: A Tool Use Perspective","primary_cat":"cs.NE","submitted_at":"2019-07-03T17:35:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Proposes a tool-use inspired framework with multiple test sets to measure specified types of generalization in RL.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1907.01749","ref_index":33,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Polyphone Disambiguation for Mandarin Chinese Using Conditional Neural Network with Multi-level Embedding Features","primary_cat":"cs.CL","submitted_at":"2019-07-03T05:50:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A conditional neural network using bidirectional RNN sentence encoding and multi-level word/sentence embeddings reaches 94.69% accuracy on a public Mandarin polyphone dataset.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1907.01022","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Rare Disease Detection by Sequence Modeling with Generative Adversarial Networks","primary_cat":"cs.LG","submitted_at":"2019-07-01T19:18:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A GAN-boosted RNN model reaches 0.56 PR-AUC for rare EPI detection on 1.8 million patients and outperforms benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1906.10910","ref_index":5,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Creating A Neural Pedagogical Agent by Jointly Learning to Review and Assess","primary_cat":"cs.LG","submitted_at":"2019-06-26T08:37:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Bidirectional RNN with attention models real-time user knowledge from question-response sequences to predict correctness, outperforming baselines especially for new users on a large TOEIC mobile app dataset.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1906.09808","ref_index":4,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Recurrent Adversarial Service Times","primary_cat":"stat.ML","submitted_at":"2019-06-24T09:31:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RNN for arrivals paired with recurrent GAN for service times to model queuing dynamics without assuming specific inter-event distributions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1906.09510","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Learning Belief Representations for Imitation Learning in POMDPs","primary_cat":"cs.LG","submitted_at":"2019-06-22T21:40:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"BMIL learns belief modules jointly with policies for GAIL-style imitation learning in POMDPs, outperforming separate training and standard GAIL on continuous control tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1906.09427","ref_index":1,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Alchemy: A Quantum Chemistry Dataset for Benchmarking AI Models","primary_cat":"cs.LG","submitted_at":"2019-06-22T10:27:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Alchemy releases a new quantum chemistry dataset of 119k molecules and 12 properties plus GNN benchmarks to support AI model development for molecular science.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1511.06939","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Session-based Recommendations with Recurrent Neural Networks","primary_cat":"cs.LG","submitted_at":"2015-11-21T23:42:59+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"RNNs with ranking loss outperform item-to-item baselines for session-based recommendations on two datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}