# Research Papers — 2026-01 · bundle page 154 of 189

> Full-text markdown of papers #7651–7700 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-01/page/154
[← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/153.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-01/page/155.md)

### Hybrid Federated Learning for Noise-Robust Training

Authors: Yongjun Kim, Hyeongjun Park, Hwanjin Kim, Junil Choi
Published: 2026-01-08
Categories: cs.LG, cs.AI, cs.IT, eess.SP
arXiv: https://arxiv.org/abs/2601.04483v1

Federated learning (FL) and federated distillation (FD) are distributed learning paradigms that train UE models with enhanced privacy, each offering different trade-offs between noise robustness and learning speed. To mitigate their respective weaknesses, we propose a hybrid federated learning (HFL) framework in which each user equipment (UE) transmits either gradients or logits, and the base station (BS) selects the per-round weights of FL and FD updates. We derive convergence of HFL framework and introduce two methods to exploit degrees of freedom (DoF) in HFL, which are (i) adaptive UE clustering via Jenks optimization and (ii) adaptive weight selection via a damped Newton method. Numerical results show that HFL achieves superior test accuracy at low SNR when both DoF are exploited.

---

### When Models Manipulate Manifolds: The Geometry of a Counting Task

Authors: Wes Gurnee, Emmanuel Ameisen, Isaac Kauvar, Julius Tarng, Adam Pearce, Chris Olah, Joshua Batson
Published: 2026-01-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.04480v1

Language models can perceive visual properties of text despite receiving only sequences of tokens-we mechanistically investigate how Claude 3.5 Haiku accomplishes one such task: linebreaking in fixed-width text. We find that character counts are represented on low-dimensional curved manifolds discretized by sparse feature families, analogous to biological place cells. Accurate predictions emerge from a sequence of geometric transformations: token lengths are accumulated into character count manifolds, attention heads twist these manifolds to estimate distance to the line boundary, and the decision to break the line is enabled by arranging estimates orthogonally to create a linear decision boundary. We validate our findings through causal interventions and discover visual illusions--character sequences that hijack the counting mechanism. Our work demonstrates the rich sensory processing of early layers, the intricacy of attention algorithms, and the importance of combining feature-based and geometric views of interpretability.

---

### Prediction of Cellular Malignancy Using Electrical Impedance Signatures and Supervised Machine Learning

Authors: Shadeeb Hossain
Published: 2026-01-08
Categories: eess.SP, cs.LG
arXiv: https://arxiv.org/abs/2601.04478v4

Bioelectrical properties of cells such as relative permittivity, conductivity, and characteristic time constants vary significantly between healthy and malignant cells across different frequencies. These distinctions provide a promising foundation for diagnostic and classification applications. This study systematically reviewed 20 scholarly articles to compile 535 datasets of quantitative bioelectric parameters in the kHz-MHz frequency range and evaluated their utility in predictive modeling. Three supervised machine learning algorithms- Random Forest (RF), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) were implemented and tuned using key hyperparameters to assess classification performance. In the second stage, a physics informed framework was incorporated to derive additional dielectric descriptors such as imaginary permittivity, loss tangent and charge relaxation time from the measured parameters. Random Forest based feature importance analysis was employed to identify the most discriminative dielectric parameters influencing the classification process. The results indicate that dielectric loss related parameters, particularly imaginary permittivity and conductivity, contribute significantly to the classification of cellular states. While the incorporation of physics-derived features improves model interpretability and reduces overfitting tendencies, the overall classification accuracy remains comparable to models trained using primary dielectric descriptors. The proposed approach highlights the potential of physics-informed machine learning for improving the analysis of dielectric spectroscopy data in the biomedical diagnostics.

---

### Computational Compliance for AI Regulation: Blueprint for a New Research Domain

Authors: Bill Marino, Nicholas D. Lane
Published: 2026-01-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.04474v1

The era of AI regulation (AIR) is upon us. But AI systems, we argue, will not be able to comply with these regulations at the necessary speed and scale by continuing to rely on traditional, analogue methods of compliance. Instead, we posit that compliance with these regulations will only realistically be achieved computationally: that is, with algorithms that run across the life cycle of an AI system, automatically steering it toward AIR compliance in the face of dynamic conditions. Yet despite their (we would argue) inevitability, the research community has yet to specify exactly how these algorithms for computational AIR compliance should behave - or how we should benchmark their performance. To fill these gaps, we specify a set of design goals for such algorithms. In addition, we specify a benchmark dataset that can be used to quantitatively measure whether individual algorithms satisfy these design goals. By delivering this blueprint, we hope to give shape to an important but uncrystallized new domain of research - and, in doing so, incite necessary investment in it.

---

### Convergence Rates for Learning Pseudo-Differential Operators

Authors: Jiaheng Chen, Daniel Sanz-Alonso
Published: 2026-01-08
Categories: math.ST, cs.LG, math.NA, stat.ML
arXiv: https://arxiv.org/abs/2601.04473v1

This paper establishes convergence rates for learning elliptic pseudo-differential operators, a fundamental operator class in partial differential equations and mathematical physics. In a wavelet-Galerkin framework, we formulate learning over this class as a structured infinite-dimensional regression problem with multiscale sparsity. Building on this structure, we propose a sparse, data- and computation-efficient estimator, which leverages a novel matrix compression scheme tailored to the learning task and a nested-support strategy to balance approximation and estimation errors. In addition to obtaining convergence rates for the estimator, we show that the learned operator induces an efficient and stable Galerkin solver whose numerical error matches its statistical accuracy. Our results therefore contribute to bringing together operator learning, data-driven solvers, and wavelet methods in scientific computing.

---

### SampoNLP: A Self-Referential Toolkit for Morphological Analysis of Subword Tokenizers

Authors: Iaroslav Chelombitko, Ekaterina Chelombitko, Aleksey Komissarov
Published: 2026-01-08
Categories: cs.CL, cs.IR, cs.LG
arXiv: https://arxiv.org/abs/2601.04469v1

The quality of subword tokenization is critical for Large Language Models, yet evaluating tokenizers for morphologically rich Uralic languages is hampered by the lack of clean morpheme lexicons. We introduce SampoNLP, a corpus-free toolkit for morphological lexicon creation using MDL-inspired Self-Referential Atomicity Scoring, which filters composite forms through internal structural cues - suited for low-resource settings. Using the high-purity lexicons generated by SampoNLP for Finnish, Hungarian, and Estonian, we conduct a systematic evaluation of BPE tokenizers across a range of vocabulary sizes (8k-256k). We propose a unified metric, the Integrated Performance Score (IPS), to navigate the trade-off between morpheme coverage and over-splitting. By analyzing the IPS curves, we identify the "elbow points" of diminishing returns and provide the first empirically grounded recommendations for optimal vocabulary sizes (k) in these languages. Our study not only offers practical guidance but also quantitatively demonstrates the limitations of standard BPE for highly agglutinative languages. The SampoNLP library and all generated resources are made publicly available: https://github.com/AragonerUA/SampoNLP

---

### Concept Tokens: Learning Behavioral Embeddings Through Concept Definitions

Authors: Ignacio Sastre, Aiala Rosá
Published: 2026-01-08
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.04465v1

We propose Concept Tokens, a lightweight method that adds a new special token to a pretrained LLM and learns only its embedding from multiple natural language definitions of a target concept, where occurrences of the concept are replaced by the new token. The LLM is kept frozen and the embedding is optimized with the standard language-modeling objective. We evaluate Concept Tokens in three settings. First, we study hallucinations in closed-book question answering on HotpotQA and find a directional effect: negating the hallucination token reduces hallucinated answers mainly by increasing abstentions, whereas asserting it increases hallucinations and lowers precision. Second, we induce recasting, a pedagogical feedback strategy for second language teaching, and observe the same directional effect. Moreover, compared to providing the full definitional corpus in-context, concept tokens better preserve compliance with other instructions (e.g., asking follow-up questions). Finally, we include a qualitative study with the Eiffel Tower and a fictional "Austral Tower" to illustrate what information the learned embeddings capture and where their limitations emerge. Overall, Concept Tokens provide a compact control signal learned from definitions that can steer behavior in frozen LLMs.

---

### Beyond Static Summarization: Proactive Memory Extraction for LLM Agents

Authors: Chengyuan Yang, Zequn Sun, Wei Wei, Wei Hu
Published: 2026-01-08
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.04463v2

Memory management is vital for LLM agents in long-term and personalized interactions. Most previous work studies how to retrieve and use memory, but pays less attention to how memory is extracted. We find two main limitations in existing methods. First, extraction is "ahead-of-time": the agent saves information before it knows future tasks. A single summary prompt often mixes details, events, and relations, so useful information is lost. Second, extraction is usually one-off. Without verification, errors and hallucinations may stay in memory for a long time. To address these limitations, we propose ProMem, a proactive memory extraction framework. It separates details, events, and relations, and uses different extraction strategies for each type. It also checks completeness to recover missed events and verifies facts at the atomic level to reduce hallucinations. Experiments show that ProMem improves memory completeness and QA accuracy, while keeping a good balance between quality and token cost.

---

### Meta-probabilistic Modeling

Authors: Kevin Zhang, Yixin Wang
Published: 2026-01-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.04462v3

Probabilistic graphical models (PGMs) are widely used to discover latent structure in data, but their success hinges on selecting an appropriate model design. In practice, model specification is difficult and often requires iterative trial-and-error. This challenge arises because classical PGMs typically operate on individual datasets. In this work, we consider settings involving collections of related datasets and propose meta-probabilistic modeling (MPM) to learn the generative model structure itself. MPM uses a hierarchical formulation in which global components encode shared patterns across datasets, while local parameters capture dataset-specific latent structure. For scalable learning and inference, we derive a tractable VAE-inspired surrogate objective together with a bi-level optimization algorithm. Our methodology supports a broad class of expressive probabilistic models and has connections to existing architectures, such as Slot Attention. Experiments on object-centric representation learning and sequential text modeling demonstrate that MPM effectively adapts generative models to data while recovering meaningful latent representations.

---

### Users Mispredict Their Own Preferences for AI Writing Assistance

Authors: Vivian Lai, Zana Buçinca, Nil-Jana Akpinar, Mo Houtti, Hyeonsu B. Kang, Kevin Chian, Namjoon Suh, Alex C. Williams
Published: 2026-01-08
Categories: cs.CL, cs.HC
arXiv: https://arxiv.org/abs/2601.04461v1

Proactive AI writing assistants need to predict when users want drafting help, yet we lack empirical understanding of what drives preferences. Through a factorial vignette study with 50 participants making 750 pairwise comparisons, we find compositional effort dominates decisions ($ρ= 0.597$) while urgency shows no predictive power ($ρ\approx 0$). More critically, users exhibit a striking perception-behavior gap: they rank urgency first in self-reports despite it being the weakest behavioral driver, representing a complete preference inversion. This misalignment has measurable consequences. Systems designed from users' stated preferences achieve only 57.7\% accuracy, underperforming even naive baselines, while systems using behavioral patterns reach significantly higher 61.3\% ($p < 0.05$). These findings demonstrate that relying on user introspection for system design actively misleads optimization, with direct implications for proactive natural language generation (NLG) systems.

---

### Using Large Language Models to Detect Socially Shared Regulation of Collaborative Learning

Authors: Jiayi Zhang, Conrad Borchers, Clayton Cohn, Namrata Srivastava, Caitlin Snyder, Siyuan Guo, Ashwin T S, Naveeduddin Mohammed, Haley Noh, Gautam Biswas
Published: 2026-01-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.04458v1

The field of learning analytics has made notable strides in automating the detection of complex learning processes in multimodal data. However, most advancements have focused on individualized problem-solving instead of collaborative, open-ended problem-solving, which may offer both affordances (richer data) and challenges (low cohesion) to behavioral prediction. Here, we extend predictive models to automatically detect socially shared regulation of learning (SSRL) behaviors in collaborative computational modeling environments using embedding-based approaches. We leverage large language models (LLMs) as summarization tools to generate task-aware representations of student dialogue aligned with system logs. These summaries, combined with text-only embeddings, context-enriched embeddings, and log-derived features, were used to train predictive models. Results show that text-only embeddings often achieve stronger performance in detecting SSRL behaviors related to enactment or group dynamics (e.g., off-task behavior or requesting assistance). In contrast, contextual and multimodal features provide complementary benefits for constructs such as planning and reflection. Overall, our findings highlight the promise of embedding-based models for extending learning analytics by enabling scalable detection of SSRL behaviors, ultimately supporting real-time feedback and adaptive scaffolding in collaborative learning environments that teachers value.

---

### Categorical Belief Propagation: Sheaf-Theoretic Inference via Descent and Holonomy

Authors: Enrique ter Horst, Sridhar Mahadevan, Juan Diego Zambrano
Published: 2026-01-08
Categories: cs.AI, math.CT
arXiv: https://arxiv.org/abs/2601.04456v1

We develop a categorical foundation for belief propagation on factor graphs. We construct the free hypergraph category \(\Syn_Σ\) on a typed signature and prove its universal property, yielding compositional semantics via a unique functor to the matrix category \(\cat{Mat}_R\). Message-passing is formulated using a Grothendieck fibration \(\int\Msg \to \cat{FG}_Σ\) over polarized factor graphs, with schedule-indexed endomorphisms defining BP updates. We characterize exact inference as effective descent: local beliefs form a descent datum when compatibility conditions hold on overlaps. This framework unifies tree exactness, junction tree algorithms, and loopy BP failures under sheaf-theoretic obstructions. We introduce HATCC (Holonomy-Aware Tree Compilation), an algorithm that detects descent obstructions via holonomy computation on the factor nerve, compiles non-trivial holonomy into mode variables, and reduces to tree BP on an augmented graph. Complexity is \(O(n^2 d_{\max} + c \cdot k_{\max} \cdot δ_{\max}^3 + n \cdot δ_{\max}^2)\) for \(n\) factors and \(c\) fundamental cycles. Experimental results demonstrate exact inference with significant speedup over junction trees on grid MRFs and random graphs, along with UNSAT detection on satisfiability instances.

---

### Re-Rankers as Relevance Judges

Authors: Chuan Meng, Jiqun Liu, Mohammad Aliannejadi, Fengran Mo, Jeff Dalton, Maarten de Rijke
Published: 2026-01-08
Categories: cs.IR, cs.AI, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2601.04455v1

Using large language models (LLMs) to predict relevance judgments has shown promising results. Most studies treat this task as a distinct research line, e.g., focusing on prompt design for predicting relevance labels given a query and passage. However, predicting relevance judgments is essentially a form of relevance prediction, a problem extensively studied in tasks such as re-ranking. Despite this potential overlap, little research has explored reusing or adapting established re-ranking methods to predict relevance judgments, leading to potential resource waste and redundant development. To bridge this gap, we reproduce re-rankers in a re-ranker-as-relevance-judge setup. We design two adaptation strategies: (i) using binary tokens (e.g., "true" and "false") generated by a re-ranker as direct judgments, and (ii) converting continuous re-ranking scores into binary labels via thresholding. We perform extensive experiments on TREC-DL 2019 to 2023 with 8 re-rankers from 3 families, ranging from 220M to 32B, and analyse the evaluation bias exhibited by re-ranker-based judges. Results show that re-ranker-based relevance judges, under both strategies, can outperform UMBRELA, a state-of-the-art LLM-based relevance judge, in around 40% to 50% of the cases; they also exhibit strong self-preference towards their own and same-family re-rankers, as well as cross-family bias.

---

### Retrieval Heads are Dynamic

Authors: Yuping Lin, Zitao Li, Yue Xing, Pengfei He, Yingqian Cui, Yaliang Li, Bolin Ding, Jingren Zhou, Jiliang Tang
Published: 2026-01-07
Categories: cs.CL
arXiv: https://arxiv.org/abs/2602.11162v2

Recent studies have identified "retrieval heads" in Large Language Models (LLMs) responsible for extracting information from input contexts. However, prior works largely rely on static statistics aggregated across datasets, identifying heads that perform retrieval on average. This perspective overlooks the fine-grained temporal dynamics of autoregressive generation. In this paper, we investigate retrieval heads from a dynamic perspective. Through extensive analysis, we establish three core claims: (1) Dynamism: Retrieval heads vary dynamically across timesteps; (2) Irreplaceability: Dynamic retrieval heads are specific at each timestep and cannot be effectively replaced by static retrieval heads; and (3) Correlation: The model's hidden state encodes a predictive signal for future retrieval head patterns, indicating an internal planning mechanism. We validate these findings on the Needle-in-a-Haystack task and a multi-hop QA task, and quantify the differences on the utility of dynamic and static retrieval heads in a Dynamic Retrieval-Augmented Generation framework. Our study provides new insights into the internal mechanisms of LLMs.

---

### Attention Isn't All You Need for Emotion Recognition:Domain Features Outperform Transformers on the EAV Dataset

Authors: Anmol Guragain
Published: 2026-01-07
Categories: cs.LG, cs.CV, cs.SD, eess.AS
arXiv: https://arxiv.org/abs/2601.22161v2

We present a systematic study of multimodal emotion recognition using the EAV dataset, investigating whether complex attention mechanisms improve performance on small datasets. We implement three model categories: baseline transformers (M1), novel factorized attention mechanisms (M2), and improved CNN baselines (M3). Our experiments show that sophisticated attention mechanisms consistently underperform on small datasets. M2 models achieved 5 to 13 percentage points below baselines due to overfitting and destruction of pretrained features. In contrast, simple domain-appropriate modifications proved effective: adding delta MFCCs to the audio CNN improved accuracy from 61.9% to 65.56% (+3.66pp), while frequency-domain features for EEG achieved 67.62% (+7.62pp over the paper baseline). Our vision transformer baseline (M1) reached 75.30%, exceeding the paper's ViViT result (74.5%) through domain-specific pretraining, and vision delta features achieved 72.68% (+1.28pp over the paper CNN). These findings demonstrate that for small-scale emotion recognition, domain knowledge and proper implementation outperform architectural complexity.

---

### Lowest Span Confidence: Zero-Shot Hallucination Detection from a Single LLM Response

Authors: Yitong Qiao, Licheng Pan, Yu Mi, Lei Liu, Yue Shen, Jian Wang, Jinjie Gu, Fei Sun, Zhixuan Chu
Published: 2026-01-07
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.19918v2

Hallucinations in Large Language Models (LLMs), i.e., plausible but non-factual generations, pose a significant challenge to reliable deployment in high-stakes environments. However, many existing hallucination detectors require expensive repeated sampling for consistency checks or access to model-internal states unavailable in common API-based scenarios. To this end, we propose an efficient zero-shot metric called Lowest Span Confidence (LSC) for hallucination detection under minimal resource assumptions. Concretely, LSC evaluates the local confidence of adjacent complete-word spans. By selecting the lowest aggregated confidence across neighboring words whose token widths can vary, LSC captures localized uncertainty associated with factual inconsistency. This boundary-aligned smoothing reduces the global dilution of perplexity and the sensitivity of minimum token probability to isolated noise. Our main evaluation spans four model families {Llama-2, Qwen2.5, Gemma-2, Mistral} and seven benchmarks {NQ, TriviaQA, SQuAD, CoQA, HotpotQA, RAGTruth, FELM}. Additional analyses examine word reconstruction, span width, and the role of adjacency in preserving local confidence. Across these settings, LSC is competitive with methods that use multiple responses or model-internal information while requiring only one response and its output token probabilities, without training a separate detector or using an auxiliary model.

---

### PILOT: Planning via Internalized Latent Optimization Trajectories for Large Language Models

Authors: Haoyu Zheng, Yun Zhu, Yuqian Yuan, Bo Yuan, Wenqiao Zhang, Siliang Tang, Jun Xiao
Published: 2026-01-07
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.19917v2

Strategic planning is critical for multi-step reasoning, yet compact Large Language Models (LLMs) often lack the capacity to formulate global strategies, leading to error propagation in long-horizon tasks. Our analysis reveals that LLMs possess latent reasoning capabilities that can be unlocked when conditioned on explicit plans from a teacher model; however, runtime reliance on external guidance is often impractical due to latency and availability constraints. To bridge this gap, we propose PILOT (Planning via Internalized Latent Optimization Trajectories), a non-invasive framework designed to internalize the strategic oversight of large models into intrinsic Latent Guidance. Instead of altering backbone weights, PILOT employs a lightweight Hyper-Network to synthesize a query-conditioned Latent Guidance vector. This vector acts as an internal steering mechanism, guiding the model's representations toward optimal reasoning paths. Extensive experiments on mathematical and coding benchmarks demonstrate that PILOT effectively stabilizes reasoning trajectories, consistently outperforming strong baselines (e.g., +8.9% on MATH500) with negligible inference latency.

---

### PaperAudit-Bench: Benchmarking Error Detection in Research Papers for Critical Automated Peer Review

Authors: Songjun Tu, Yiwen Ma, Jiahao Lin, Qichao Zhang, Xiangyuan Lan, Junfeng. Li, Nan Xu, Linjing Li, Dongbin Zhao
Published: 2026-01-07
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.19916v1

Large language models can generate fluent peer reviews, yet their assessments often lack sufficient critical rigor when substantive issues are subtle and distributed across a paper. In this paper, we introduce PaperAudit-Bench, which consists of two components: (1) PaperAudit-Dataset, an error dataset covering both errors identifiable within individual sections and those requiring cross-section reasoning, designed for controlled evaluation under long-context settings; and (2) PaperAudit-Review, an automated review framework that integrates structured error detection with evidence-aware review generation to support critical assessment. Experiments on PaperAudit-Bench reveal large variability in error detectability across models and detection depths, highlighting the difficulty of identifying such errors under long-context settings. Relative to representative automated reviewing baselines, incorporating explicit error detection into the review workflow produces systematically stricter and more discriminative evaluations, demonstrating its suitability for peer review. Finally, we show that the dataset supports training lightweight LLM detectors via SFT and RL, enabling effective error detection at reduced computational cost.

---

### Modeling Next-Token Prediction as Left-Nested Intuitionistic Implication

Authors: Paul Tarau
Published: 2026-01-07
Categories: cs.CL, cs.AI, cs.LO
arXiv: https://arxiv.org/abs/2601.19915v1

We introduce the \emph{Arrow Language Model}, a neural architecture derived from an intuitionistic-logic interpretation of next-token prediction. Instead of representing tokens as additive embeddings mixed by attention, we encode a prefix as a \emph{left-nested implication chain} whose structure preserves order through non-commutative composition. Next-token prediction corresponds to \emph{modus ponens}, and sequence processing becomes constructive proof extension under the Curry--Howard correspondence. Our Prolog-based specialized theorem provers validate fundamental properties of the neural models, among which relations between commutative vs. non-commutative sequencing and single-token vs. multi-token prediction choices. We show that a neural architecture equivalent to multiplicative RNNs arises naturally from a proof-theoretic interpretation of next-token prediction as nested intuitionistic implication, we present a practical low-rank neural realization and position the model relative to Transformers and state-space models. Keywords: logic-based derivation of neural architectures, intuitionistic implicational logic, token-as-operator neural models, state-space models, alternatives to transformer-based foundational models.

---

### The Voice of Equity: A Systematic Evaluation of Bias Mitigation Techniques for Speech-Based Cognitive Impairment Detection Across Architectures and Demographics

Authors: Yasaman Haghbin, Sina Rashidi, Ali Zolnour, Maryam Zolnoori
Published: 2026-01-07
Categories: eess.AS, cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2601.16989v1

Speech-based detection of cognitive impairment offers a scalable, non-invasive screening, yet algorithmic bias across demographic and linguistic subgroups remains critically underexplored. We present the first comprehensive fairness analysis framework for speech-based multi-class cognitive impairment detection, systematically evaluating bias mitigation across architectures, and demographic subgroups. We developed two transformer-based architectures, SpeechCARE-AGF and Whisper-LWF-LoRA, on the multilingual NIA PREPARE Challenge dataset. Unlike prior work that typically examines single mitigation techniques, we compared pre-processing, in-processing, and post-processing approaches, assessing fairness via Equality of Opportunity and Equalized Odds across gender, age, education, and language. Both models achieved strong performance (F1: SpeechCARE-AGF 70.87, Whisper-LWF-LoRA 71.46) but exhibited substantial fairness disparities. Adults >=80 showed lower sensitivity versus younger groups; Spanish speakers demonstrated reduced TPR versus English speakers. Mitigation effectiveness varied by architecture: oversampling improved SpeechCARE-AGF for older adults (80+ TPR: 46.19%=>49.97%) but minimally affected Whisper-LWF-LoRA. This study addresses a critical healthcare AI gap by demonstrating that architectural design fundamentally shapes bias patterns and mitigation effectiveness. Adaptive fusion mechanisms enable flexible responses to data interventions, while frequency reweighting offers robust improvements across architectures. Our findings establish that fairness interventions must be tailored to both model architecture and demographic characteristics, providing a systematic framework for developing equitable speech-based screening tools essential for reducing diagnostic disparities in cognitive healthcare.

---

### Towards Latent Diffusion Suitable For Text

Authors: Nesta Midavaine, Christian A. Naesseth, Grigory Bartosh
Published: 2026-01-07
Categories: cs.CL, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2601.16220v1

Language diffusion models aim to improve sampling speed and coherence over autoregressive LLMs. We introduce Neural Flow Diffusion Models for language generation, an extension of NFDM that enables the straightforward application of continuous diffusion models to discrete state spaces. NFDM learns a multivariate forward process from the data, ensuring that the forward process and generative trajectory are a good fit for language modeling. Our model substantially reduces the likelihood gap with autoregressive models of the same size, while achieving sample quality comparable to that of previous latent diffusion models.

---

### Intelligence Degradation in Long-Context LLMs: Critical Threshold Determination via Natural Length Distribution Analysis

Authors: Weiwei Wang, Jiyong Min, Weijie Zou
Published: 2026-01-07
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.15300v1

Large Language Models (LLMs) exhibit catastrophic performance degradation when processing contexts approaching certain critical thresholds, even when information remains relevant. This intelligence degradation-defined as over 30% drop in task performance-severely limits long-context applications. This degradation shows a common pattern: models maintain strong performance up to a critical threshold, then collapse catastrophically. We term this shallow long-context adaptation-models adapt for short to medium contexts but fail beyond critical thresholds. This paper presents three contributions: (1) Natural Length Distribution Analysis: We use each sample's natural token length without truncation or padding, providing stronger causal evidence that degradation results from context length itself. (2) Critical Threshold Determination: Through experiments on a mixed dataset (1,000 samples covering 5%-95% of context length), we identify the critical threshold for Qwen2.5-7B at 40-50% of maximum context length, where F1 scores drop from 0.55-0.56 to 0.3 (45.5% degradation), using five-method cross-validation. (3) Unified Framework: We consolidate shallow adaptation, explaining degradation patterns and providing a foundation for mitigation strategies. This work provides the first systematic characterization of intelligence degradation in open-source Qwen models, offering practical guidance for deploying LLMs in long-context scenarios.

---

### MALTopic: Multi-Agent LLM Topic Modeling Framework

Authors: Yash Sharma
Published: 2026-01-07
Categories: cs.CL, cs.IR, cs.MA
arXiv: https://arxiv.org/abs/2601.15299v1

Topic modeling is a crucial technique for extracting latent themes from unstructured text data, particularly valuable in analyzing survey responses. However, traditional methods often only consider free-text responses and do not natively incorporate structured or categorical survey responses for topic modeling. And they produce abstract topics, requiring extensive human interpretation. To address these limitations, we propose the Multi-Agent LLM Topic Modeling Framework (MALTopic). This framework decomposes topic modeling into specialized tasks executed by individual LLM agents: an enrichment agent leverages structured data to enhance textual responses, a topic modeling agent extracts latent themes, and a deduplication agent refines the results. Comparative analysis on a survey dataset demonstrates that MALTopic significantly improves topic coherence, diversity, and interpretability compared to LDA and BERTopic. By integrating structured data and employing a multi-agent approach, MALTopic generates human-readable topics with enhanced contextual relevance, offering a more effective solution for analyzing complex survey data.

---

### Toward Youth-Centered Privacy-by-Design in Smart Devices: A Systematic Review

Authors: Molly Campbell, Mohamad Sheikho Al Jasem, Ajay Kumar Shrestha
Published: 2026-01-07
Categories: cs.CY, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2601.11598v2

This literature review evaluates privacy-by-design frameworks, tools, and policies intended to protect youth in AI-enabled smart devices using a PRISMA-guided workflow. Sources from major academic and grey-literature repositories from the past decade were screened. The search identified 2,216 records; after deduplication and screening, 645 articles underwent eligibility assessment, and 122 were included for analysis. The corpus was organized along three thematic categories: technical solutions, policy/regulatory measures, and education/awareness strategies. Findings reveal that while technical interventions such as on-device processing, federated learning, and lightweight encryption significantly reduce data exposure, their adoption remains limited. Policy frameworks, including the EU's GDPR, the UK Age-Appropriate Design Code, and Canada's PIPEDA, provide important baselines but are hindered by gaps in enforcement and age-appropriate design obligations, while educational initiatives are rarely integrated systematically into curricula. Overall, the corpus skews toward technical solutions (67%) relative to policy (21%) and education (12%), indicating an implementation gap outside the technical domain. To address these challenges, we recommend a multi-stakeholder model in which policymakers, manufacturers, and educators co-develop inclusive, transparent, and context-sensitive privacy ecosystems. This work advances discourse on youth data protection by offering empirically grounded insights and actionable recommendations for the design of ethical, privacy-preserving AI systems tailored to young users.

---

### SSC-UNet: UNet with Self-Supervised Contrastive Learning for Phonocardiography Noise Reduction

Authors: Lizy Abraham, Siobhan Coughlan, Kritika Rajain, Changhong Li, Saji Philip, Adam James
Published: 2026-01-07
Categories: eess.SP, cs.LG
arXiv: https://arxiv.org/abs/2601.10735v1

Congenital Heart Disease (CHD) remains a significant global health concern affecting approximately 1\% of births worldwide. Phonocardiography has emerged as a supplementary tool to diagnose CHD cost-effectively. However, the performance of these diagnostic models highly depends on the quality of the phonocardiography, thus, noise reduction is particularly critical. Supervised UNet effectively improves noise reduction capabilities, but limited clean data hinders its application. The complex time-frequency characteristics of phonocardiography further complicate finding the balance between effectively removing noise and preserving pathological features. In this study, we proposed a self-supervised phonocardiography noise reduction model based on Noise2Noise to enable training without clean data. Augmentation and contrastive learning are applied to enhance its performance. We obtained an average SNR of 12.98 dB after filtering under 10~dB of hospital noise. Classification sensitivity after filtering was improved from 27\% to 88\%, indicating its promising pathological feature retention capabilities in practical noisy environments.

---

### Reinforced Linear Genetic Programming

Authors: Urmzd Mukhammadnaim
Published: 2026-01-07
Categories: cs.NE, cs.AI
arXiv: https://arxiv.org/abs/2601.09736v1

Linear Genetic Programming (LGP) is a powerful technique that allows for a variety of problems to be solved using a linear representation of programs. However, there still exists some limitations to the technique, such as the need for humans to explicitly map registers to actions. This thesis proposes a novel approach that uses Q-Learning on top of LGP, Reinforced Linear Genetic Programming (RLGP) to learn the optimal register-action assignments. In doing so, we introduce a new framework "linear-gp" written in memory-safe Rust that allows for extensive experimentation for future works.

---

### Adaptive Trust Metrics for Multi-LLM Systems: Enhancing Reliability in Regulated Industries

Authors: Tejaswini Bollikonda
Published: 2026-01-07
Categories: cs.SE, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2601.08858v1

Large Language Models (LLMs) are increasingly deployed in sensitive domains such as healthcare, finance, and law, yet their integration raises pressing concerns around trust, accountability, and reliability. This paper explores adaptive trust metrics for multi LLM ecosystems, proposing a framework for quantifying and improving model reliability under regulated constraints. By analyzing system behaviors, evaluating uncertainty across multiple LLMs, and implementing dynamic monitoring pipelines, the study demonstrates practical pathways for operational trustworthiness. Case studies from financial compliance and healthcare diagnostics illustrate the applicability of adaptive trust metrics in real world settings. The findings position adaptive trust measurement as a foundational enabler for safe and scalable AI adoption in regulated industries.

---

### Decoder Generates Manufacturable Structures: A Framework for 3D-Printable Object Synthesis

Authors: Abhishek Kumar
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.08015v1

This paper presents a novel decoder-based approach for generating manufacturable 3D structures optimized for additive manufacturing. We introduce a deep learning framework that decodes latent representations into geometrically valid, printable objects while respecting manufacturing constraints such as overhang angles, wall thickness, and structural integrity. The methodology demonstrates that neural decoders can learn complex mapping functions from abstract representations to valid 3D geometries, producing parts with significantly improved manufacturability compared to naive generation approaches. We validate the approach on diverse object categories and demonstrate practical 3D printing of decoder-generated structures.

---

### A Unified Attention U-Net Framework for Cross-Modality Tumor Segmentation in MRI and CT

Authors: Nishan Rai, Pushpa R. Dahal
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.06187v1

This study presents a unified Attention U-Net architecture trained jointly on MRI (BraTS 2021) and CT (LIDC-IDRI) datasets to investigate the generalizability of a single model across diverse imaging modalities and anatomical sites. Our proposed pipeline incorporates modality-harmonized preprocessing, attention-gated skip connections, and a modality-aware Focal Tversky loss function. To the best of our knowledge, this study is among the first to evaluate a single Attention U-Net trained simultaneously on separate MRI (BraTS) and CT (LIDC-IDRI) tumor datasets, without relying on modality-specific encoders or domain adaptation. The unified model demonstrates competitive performance in terms of Dice coefficient, IoU, and AUC on both domains, thereby establishing a robust and reproducible baseline for future research in cross-modality tumor segmentation.

---

### Time-Series Anomaly Classification for Launch Vehicle Propulsion Systems: Fast Statistical Detectors Enhancing LSTM Accuracy and Data Quality

Authors: Sean P. Engelstad, Sameul R. Darr, Matthew Taliaferro, Vinay K. Goyal
Published: 2026-01-07
Categories: cs.LG, stat.AP, stat.ML
arXiv: https://arxiv.org/abs/2601.06186v1

Supporting Go/No-Go decisions prior to launch requires assessing real-time telemetry data against redline limits established during the design qualification phase. Family data from ground testing or previous flights is commonly used to detect initiating failure modes and their timing; however, this approach relies heavily on engineering judgment and is more error-prone for new launch vehicles. To address these limitations, we utilize Long-Term Short-Term Memory (LSTM) networks for supervised classification of time-series anomalies. Although, initial training labels derived from simulated anomaly data may be suboptimal due to variations in anomaly strength, anomaly settling times, and other factors. In this work, we propose a novel statistical detector based on the Mahalanobis distance and forward-backward detection fractions to adjust the supervised training labels. We demonstrate our method on digital twin simulations of a ground-stage propulsion system with 20.8 minutes of operation per trial and O(10^8) training timesteps. The statistical data relabeling improved precision and recall of the LSTM classifier by 7% and 22% respectively.

---

### Attention Mechanism and Heuristic Approach: Context-Aware File Ranking Using Multi-Head Self-Attention

Authors: Pradeep Kumar Sharma, Shantanu Godbole, Sarada Prasad Jena, Hritvik Shrivastava
Published: 2026-01-07
Categories: cs.SE, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.06185v1

The identification and ranking of impacted files within software reposi-tories is a key challenge in change impact analysis. Existing deterministic approaches that combine heuristic signals, semantic similarity measures, and graph-based centrality metrics have demonstrated effectiveness in nar-rowing candidate search spaces, yet their recall plateaus. This limitation stems from the treatment of features as linearly independent contributors, ignoring contextual dependencies and relationships between metrics that characterize expert reasoning patterns. To address this limitation, we propose the application of Multi-Head Self-Attention as a post-deterministic scoring refinement mechanism. Our approach learns contextual weighting between features, dynamically adjust-ing importance levels per file based on relational behavior exhibited across candidate file sets. The attention mechanism produces context-aware adjustments that are additively combined with deterministic scores, pre-serving interpretability while enabling reasoning similar to that performed by experts when reviewing change surfaces. We focus on recall rather than precision, as false negatives (missing impacted files) are far more costly than false positives (irrelevant files that can be quickly dismissed during review). Empirical evaluation on 200 test cases demonstrates that the introduc-tion of self-attention improves Top-50 recall from approximately 62-65% to between 78-82% depending on repository complexity and structure, achiev-ing 80% recall at Top-50 files. Expert validation yields improvement from 6.5/10 to 8.6/10 in subjective accuracy alignment. This transformation bridges the reasoning capability gap between deterministic automation and expert judgment, improving recall in repository-aware effort estimation.

---

### Data-Driven Reduced-Complexity Modeling of Fluid Flows: A Community Challenge

Authors: Oliver T. Schmidt, Aaron Towne, Adrian Lozano-Duran, Scott T. M. Dawson, Ricardo Vinuesa
Published: 2026-01-07
Categories: cs.LG, nlin.CD, physics.comp-ph, physics.flu-dyn
arXiv: https://arxiv.org/abs/2601.06183v1

We introduce a community challenge designed to facilitate direct comparisons between data-driven methods for compression, forecasting, and sensing of complex aerospace flows. The challenge is organized into three tracks that target these complementary capabilities: compression (compact representations for large datasets), forecasting (predicting future flow states from a finite history), and sensing (inferring unmeasured flow states from limited measurements). Across these tracks, multiple challenges span diverse flow datasets and use cases, each emphasizing different model requirements. The challenge is open to anyone, and we invite broad participation to build a comprehensive and balanced picture of what works and where current methods fall short. To support fair comparisons, we provide standardized success metrics, evaluation tools, and baseline implementations, with one classical and one machine-learning baseline per challenge. Final assessments use blind tests on withheld data. We explicitly encourage negative results and careful analyses of limitations. Outcomes will be disseminated through an AIAA Journal Virtual Collection and invited presentations at AIAA conferences.

---

### Neuro-Symbolic Compliance: Integrating LLMs and SMT Solvers for Automated Financial Legal Analysis

Authors: Yung-Shen Hsia, Fang Yu, Jie-Hong Roland Jiang
Published: 2026-01-07
Categories: cs.AI, cs.FL, cs.LG, cs.LO
arXiv: https://arxiv.org/abs/2601.06181v1

Financial regulations are increasingly complex, hindering automated compliance-especially the maintenance of logical consistency with minimal human oversight. We introduce a Neuro-Symbolic Compliance Framework that integrates Large Language Models (LLMs) with Satisfiability Modulo Theories (SMT) solvers to enable formal verifiability and optimization-based compliance correction. The LLM interprets statutes and enforcement cases to generate SMT constraints, while the solver enforces consistency and computes the minimal factual modification required to restore legality when penalties arise. Unlike transparency-oriented methods, our approach emphasizes logic-driven optimization, delivering verifiable, legally consistent reasoning rather than post-hoc explanation. Evaluated on 87 enforcement cases from Taiwan's Financial Supervisory Commission (FSC), the system attains 86.2% correctness in SMT code generation, improves reasoning efficiency by over 100x, and consistently corrects violations-establishing a preliminary foundation for optimization-based compliance applications.

---

### MixDPO: Modeling Preference Strength for Pluralistic Alignment

Authors: Saki Imai, Pedram Heydari, Anthony Sicilia, Asteria Kaeberlein, Katherine Atwell, Malihe Alikhani
Published: 2026-01-07
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.06180v1

Preference based alignment objectives implicitly assume that all human preferences are expressed with equal strength. In practice, however, preference strength varies across individuals and contexts -- a phenomenon established in behavioral economics and discrete choice theory. This mismatch limits the ability of existing objectives to faithfully capture heterogeneous human judgments. Inspired by this literature, we introduce Mixed Logit Direct Preference Optimization (MixDPO), a generalization of Direct Preference Optimization that models variation in preference strength. MixDPO enables alignment objectives to capture heterogeneity in how strongly preferences are expressed across training examples. We evaluate MixDPO on three preference datasets using two open-weight language models. Across datasets, MixDPO improves aggregate alignment performance (+11.2 points on Pythia-2.8B) while preserving subgroup level preferences, with the largest gains appearing in settings with higher inferred preference heterogeneity. MixDPO makes preference heterogeneity explicit through learned strength distributions. We release our code for reproducibility.

---

### Performance of models for monitoring sustainable development goals from remote sensing: A three-level meta-regression

Authors: Jonas Klingwort, Nina M. Leach, Joep Burger
Published: 2026-01-07
Categories: cs.CY, cs.LG, stat.AP, stat.ME
arXiv: https://arxiv.org/abs/2601.06178v1

Machine learning (ML) is a tool to exploit remote sensing data for the monitoring and implementation of the United Nations' Sustainable Development Goals (SDGs). In this paper, we report on a meta-analysis to evaluate the performance of ML applied to remote sensing data to monitor SDGs. Specifically, we aim to 1) estimate the average performance; 2) determine the degree of heterogeneity between and within studies; and 3) assess how study features influence model performance. Using PRISMA guidelines, a search was performed across multiple academic databases to identify potentially relevant studies. A random sample of 200 was screened by three reviewers, resulting in 86 trials within 20 studies with 14 study features. Overall accuracy was the most reported performance metric. It was analyzed using double arcsine transformation and a three-level random effects model. The average overall accuracy of the best model was 0.90 (0.86, 0.92). There was considerable heterogeneity in model performance, 64% of which was between studies. The only significant feature was the prevalence of the majority class, which explained 61% of the between-study heterogeneity. None of the other thirteen features added value to the model. The most important contributions of this paper are the following two insights. 1) Overall accuracy is the most popular performance metric, yet arguably the least insightful. Its sensitivity to class imbalance makes it necessary to normalize it, which is far from common practice. 2) The field needs to standardize the reporting. Reporting of the confusion matrix for independent test sets is the most important ingredient for between-study comparisons of ML classifiers. These findings underscore the need for robust and comparable evaluation metrics in machine learning applications to ensure reliable and actionable insights for effective SDG monitoring and policy formulation.

---

### TIR-Flow: Active Video Search and Reasoning with Frozen VLMs

Authors: Hongbo Jin, Siyi Xie, Jiayu Ding, Kuanwei Lin, Ge Li
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.06176v1

While Large Video-Language Models (Video-LLMs) have achieved remarkable progress in perception, their reasoning capabilities remain a bottleneck. Existing solutions typically resort to a heavy "data engineering" paradigm-synthesizing large-scale Chain-of-Thought (CoT) datasets followed by Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). This pipeline primarily optimizes probability sampling efficiency and aligns output distributions, but fails to activate the intrinsic intelligence required for dynamic visual exploration. In this work, we propose TIR-Flow, a novel framework that shifts the paradigm from passive processing to active video searching and reasoning without additional data or parameter updating. Concretely, our framework operates through three synergistic modules: HDD decomposes complex queries into a set of verifiable sub-tasks; HAP actively directs visual attention to gather high-resolution evidence for hypothesis validation; EBA maintains a persistent workspace to accumulate and update the discovered clues for logical reasoning. Extensive experiments on seven benchmarks demonstrate that TIR-Flow significantly outperforms recent strong baselines, delivering an average performance boost of 5.9%, with gains reaching 10.5% on Egoschema. Our analysis confirms that empowering frozen VLMs with System-2-like active perception is a scalable path toward solving long-horizon video reasoning.

---

### The environmental impact of ICT in the era of data and artificial intelligence

Authors: François Rottenberg, Thomas Feys, Liesbet Van der Perre
Published: 2026-01-07
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2601.06174v1

The technology industry promotes artificial intelligence (AI) as a key enabler to solve a vast number of problems, including the environmental crisis. However, when looking at the emissions of datacenters from worldwide service providers, we observe a rapid increase aligned with the advent of AI. Some actors justify it by claiming that the increase of emissions for digital infrastructures is acceptable as it could help the decarbonization of other sectors, e.g., videoconference tools instead of taking the plane for a meeting abroad, or using AI to optimize and reduce energy consumption. With such conflicting claims and ambitions, it is unclear how the net environmental impact of AI could be quantified. The answer is prone to uncertainty for different reasons, among others: lack of transparency, interference with market expectations, lack of standardized methodology for quantifying direct and indirect impact, and the quick evolutions of models and their requirements. This report provides answers and clarifications to these different elements. Firstly, we consider the direct environmental impact of AI from a top-down approach, starting from general information and communication technologies (ICT) and then zooming in on data centers and the different phases of AI development and deployment. Secondly, a framework is introduced on how to assess both the direct and indirect impact of AI. Finally, we finish with good practices and what we can do to reduce AI impact.

---

### The Psychology of Learning from Machines: Anthropomorphic AI and the Paradox of Automation in Education

Authors: Junaid Qadir, Muhammad Mumtaz
Published: 2026-01-07
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2601.06172v2

As AI tutors enter classrooms at unprecedented speed, their deployment increasingly outpaces our grasp of the psychological and social consequences of such technology. Yet decades of research in automation psychology, human factors, and human-computer interaction provide crucial insights that remain underutilized in educational AI design. This work synthesizes four research traditions -- automation psychology, human factors engineering, HCI, and philosophy of technology -- to establish a comprehensive framework for understanding how learners psychologically relate to anthropomorphic AI tutors. We identify three persistent challenges intensified by Generative AI's conversational fluency. First, learners exhibit dual trust calibration failures -- automation bias (uncritical acceptance) and algorithm aversion (excessive rejection after errors) -- with an expertise paradox where novices overrely while experts underrely. Second, while anthropomorphic design enhances engagement, it can distract from learning and foster harmful emotional attachment. Third, automation ironies persist: systems meant to aid cognition introduce designer errors, degrade skills through disuse, and create monitoring burdens humans perform poorly. We ground this theoretical synthesis through comparative analysis of over 104,984 YouTube comments across AI-generated philosophical debates and human-created engineering tutorials, revealing domain-dependent trust patterns and strong anthropomorphic projection despite minimal cues. For engineering education, our synthesis mandates differentiated approaches: AI tutoring for technical foundations where automation bias is manageable through proper scaffolding, but human facilitation for design, ethics, and professional judgment where tacit knowledge transmission proves irreplaceable.

---

### From Individual Prompts to Collective Intelligence: Mainstreaming Generative AI in the Classroom

Authors: Junaid Qadir, Muhammad Salman Khan
Published: 2026-01-07
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2601.06171v1

Engineering classrooms are increasingly experimenting with generative AI (GenAI), but most uses remain confined to individual prompting and isolated assistance. This narrow framing risks reinforcing equity gaps and only rewarding the already privileged or motivated students. We argue instead for a shift toward collective intelligence (CI)-focused pedagogy, where GenAI acts as a catalyst for peer-to-peer learning. We implemented Generative CI (GCI) activities in two undergraduate engineering courses, engaging 140 students through thinking routines -- short, repeatable scaffolds developed by Harvard Project Zero to make thinking visible and support collaborative sense-making. Using routines such as Question Sorts and Peel the Fruit, combined with strategic AI consultation, we enabled students to externalize their reasoning, compare interpretations, and iteratively refine ideas. Our dual-pronged approach synthesizes literature from learning sciences, CI, embodied cognition, and philosophy of technology, while also empirically learning through student surveys and engagement observations. Results demonstrate that students value the combination of human collaboration with strategic AI support, recognizing risks of over-reliance while appreciating AI's role in expanding perspectives. Students identified that group work fosters deeper understanding and creative problem-solving than AI alone, with the timing of AI consultation significantly affecting learning outcomes. We offer practical implementation pathways for mainstreaming CI-focused pedagogy that cultivates deeper engagement, resilient problem-solving, and shared ownership of knowledge.

---

### Deep Joint Source-Channel Coding for Wireless Video Transmission with Asymmetric Context

Authors: Xuechen Chen, Junting Li, Chuang Chen, Hairong Lin, Yishen Li
Published: 2026-01-07
Categories: eess.IV, cs.CV
arXiv: https://arxiv.org/abs/2601.06170v1

In this paper, we propose a high-efficiency deep joint source-channel coding (JSCC) method for video transmission based on conditional coding with asymmetric context. The conditional coding-based neural video compression requires to predict the encoding and decoding conditions from the same context which includes the same reconstructed frames. However in JSCC schemes which fall into pseudo-analog transmission, the encoder cannot infer the same reconstructed frames as the decoder even a pipeline of the simulated transmission is constructed at the encoder. In the proposed method, without such a pipeline, we guide and design neural networks to learn encoding and decoding conditions from asymmetric contexts. Additionally, we introduce feature propagation, which allows intermediate features to be independently propagated at the encoder and decoder and help to generate conditions, enabling the framework to greatly leverage temporal correlation while mitigating the problem of error accumulation. To further exploit the performance of the proposed transmission framework, we implement content-adaptive coding which achieves variable bandwidth transmission using entropy models and masking mechanisms. Experimental results demonstrate that our method outperforms existing deep video transmission frameworks in terms of performance and effectively mitigates the error accumulation. By mitigating the error accumulation, our schemes can reduce the frequency of inserting intra-frame coding modes, further enhancing performance.

---

### Think Bright, Diffuse Nice: Enhancing T2I-ICL via Inductive-Bias Hint Instruction and Query Contrastive Decoding

Authors: Zhiyong Ma, Zhenpeng Li, Yuanjie Shi, Zhengping Li, Jiahao Chen, Qingyuan Chuai
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.06169v1

Text-to-Image In-Context Learning (T2I-ICL) enables customized image synthesis via interleaved text-image examples but faces two mutually reinforcing bottlenecks, compliance failure and prior-dominated hallucination, that form a vicious cycle degrading generation quality. Existing methods rely on tailored training, which limits flexibility and raises deployment costs. To address these challenges effectively, we propose TBDN, a training-free framework integrating two complementary closed-loop mechanisms: Hint Instruction (HI) and Query Contrastive Decoding (QCD). HI injects task-aware inductive bias via lightweight prompt engineering to anchor models on contextual mapping rules, thereby mitigating compliance failure. QCD adjusts the decoding distributions of language models by contrasting full-input and query-omitted distributions, suppressing prior-dominated hallucination. TBDN achieves State-of-the-Art performance on CoBSAT and Text-to-Image Fast Mini-ImageNet, with robust generalization across model backbones, prompt designs, and hyperparameters. It also maintains promising performance in concept preservation and prompt following on Dreambench++. By breaking the two bottlenecks, TBDN establishes a simple yet effective framework for efficient and reliable T2I-ICL.

---

### Analyzing the Structure of Handwritten Digits: A Comparative Study of PCA, Factor Analysis, and UMAP

Authors: Jyotiraditya Gupta
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.06168v1

Handwritten digit images lie in a high-dimensional pixel space but exhibit strong geometric and statistical structure. This paper investigates the latent organization of handwritten digits in the MNIST dataset using three complementary dimensionality reduction techniques: Principal Component Analysis (PCA), Factor Analysis (FA), and Uniform Manifold Approximation and Projection (UMAP). Rather than focusing on classification accuracy, we study how each method characterizes intrinsic dimensionality, shared variation, and nonlinear geometry. PCA reveals dominant global variance directions and enables high-fidelity reconstructions using a small number of components. FA decomposes digits into interpretable latent handwriting primitives corresponding to strokes, loops, and symmetry. UMAP uncovers nonlinear manifolds that reflect smooth stylistic transitions between digit classes. Together, these results demonstrate that handwritten digits occupy a structured low-dimensional manifold and that different statistical frameworks expose complementary aspects of this structure.

---

### Parent-Guided Adaptive Reliability (PGAR): A Behavioural Meta-Learning Framework for Stable and Trustworthy AI

Authors: Anshum Rankawat
Published: 2026-01-07
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.06167v1

Parent-Guided Adaptive Reliability (PGAR) is a lightweight behavioural meta-learning framework that adds a supervisory "parent" layer on top of a standard learner to improve stability, calibration, and recovery under disturbances. PGAR computes three reflex-level signals (incident detection, overconfidence correction, and recovery memory) and fuses them into a bounded reliability index in (0,1). This index continuously modulates the learner's effective learning rate, reducing update magnitude during instability and restoring it as reliability improves. We provide a Lyapunov-based proof sketch establishing bounded adaptation of the reliability dynamics under mild assumptions (smooth loss, descent direction, and bounded reflex outputs). Empirical evaluations on representative learning tasks show improved calibration, reduced loss variance, and faster recovery compared to standard optimizers, while retaining computational simplicity. PGAR functions as a plug-in reliability layer for existing optimization and learning pipelines, supporting interpretable reliability traces in safety-relevant settings.

---

### B-FIRE: Binning-Free Diffusion Implicit Neural Representation for Hyper-Accelerated Motion-Resolved MRI

Authors: Di Xu, Hengjie Liu, Yang Yang, Mary Feng, Jin Ning, Xin Miao, Jessica E. Scholey, Alexandra E. Hotca-cho, William C. Chen, Michael Ohliger, Martina Descovich, Huiming Dong et al.
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.06166v2

Accelerated dynamic volumetric magnetic resonance imaging (4DMRI) is essential for applications relying on motion resolution. Existing 4DMRI produces acceptable artifacts of averaged breathing phases, which can blur and misrepresent instantaneous dynamic information. Recovery of such information requires a new paradigm to reconstruct extremely undersampled non-Cartesian k-space data. We propose B-FIRE, a binning-free diffusion implicit neural representation framework for hyper-accelerated MR reconstruction capable of reflecting instantaneous 3D abdominal anatomy. B-FIRE employs a CNN-INR encoder-decoder backbone optimized using diffusion with a comprehensive loss that enforces image-domain fidelity and frequency-aware constraints. Motion binned image pairs were used as training references, while inference was performed on binning-free undersampled data. Experiments were conducted on a T1-weighted StarVIBE liver MRI cohort, with accelerations ranging from 8 spokes per frame (RV8) to RV1. B-FIRE was compared against direct NuFFT, GRASP-CS, and an unrolled CNN method. Reconstruction fidelity, motion trajectory consistency, and inference latency were evaluated.

---

### What Users Leave Unsaid: Under-Specified Queries Limit Vision-Language Models

Authors: Dasol Choi, Guijin Son, Hanwool Lee, Minhyuk Kim, Hyunwoo Ko, Teabin Lim, Ahn Eungyeol, Jungwhan Kim, Seunghyeok Hong, Youngsook Song
Published: 2026-01-07
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.06165v2

Current vision-language benchmarks predominantly feature well-structured questions with clear, explicit prompts. However, real user queries are often informal and underspecified. Users naturally leave much unsaid, relying on images to convey context. We introduce HAERAE-Vision, a benchmark of 653 real-world visual questions from Korean online communities (0.76% survival from 86K candidates), each paired with an explicit rewrite, yielding 1,306 query variants in total. Evaluating 39 VLMs, we find that even state-of-the-art models (GPT-5, Gemini 2.5 Pro) achieve under 50% on the original queries. Crucially, query explicitation alone yields 8 to 22 point improvements, with smaller models benefiting most. We further show that even with web search, under-specified queries underperform explicit queries without search, revealing that current retrieval cannot compensate for what users leave unsaid. Our findings demonstrate that a substantial portion of VLM difficulty stem from natural query under-specification instead of model capability, highlighting a critical gap between benchmark evaluation and real-world deployment.

---

### Contract2Plan: Verified Contract-Grounded Retrieval-Augmented Optimization for BOM-Aware Procurement and Multi-Echelon Inventory Planning

Authors: Sahil Agarwal
Published: 2026-01-07
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2601.06164v1

Procurement and inventory planning is governed not only by demand forecasts and bills of materials (BOMs), but also by operational terms in contracts and supplier documents (e.g., MOQs, lead times, price tiers, allocation caps, substitution approvals). LLM-based extraction can speed up structuring these terms, but extraction-only or LLM-only decision pipelines are brittle: missed clauses, unit errors, and unresolved conflicts can yield infeasible plans or silent contract violations, amplified by BOM coupling. We introduce Contract2Plan, a verified GenAI-to-optimizer pipeline that inserts a solver-based compliance gate before plans are emitted. The system retrieves clause evidence with provenance, extracts a typed constraint schema with evidence spans, compiles constraints into a BOM-aware MILP, and verifies grounding, eligibility, consistency, and feasibility using solver diagnostics, triggering targeted repair or abstention when automation is unsafe. We formalize which clause classes admit conservative repair with contract-safe feasibility guarantees and which require human confirmation. A self-contained synthetic micro-benchmark (500 instances; T=5) computed by exact enumeration under an execution model with MOQ uplift and emergency purchases shows heavy-tailed regret and nontrivial MOQ-violation incidence for extraction-only planning, motivating verification as a first-class component of contract-grounded planning systems.

---

### Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking

Authors: Kaiyuan Deng, Bo Hui, Gen Li, Jie Ji, Minghai Qin, Geng Yuan, Xiaolong Ma
Published: 2026-01-07
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2601.06163v2

The widespread adoption of text-to-image (T2I) diffusion models has raised concerns about their potential to generate copyrighted, inappropriate, or sensitive imagery. As a practical solution, machine unlearning aims to erase unwanted concepts without retraining from scratch. While most existing methods are effective for single-concept unlearning, they often struggle when removing multiple concepts, causing significant challenges in unlearning effectiveness, generation quality, and sensitivity to hyperparameters and datasets. We take a unique perspective on multi-concept unlearning by leveraging model sparsity and propose the Forget It All (FIA) framework. FIA first introduces Contrastive Concept Saliency to quantify each weight connection's contribution to a target concept. It then identifies Concept Sensitive Neurons by combining temporal and spatial information, ensuring that only neurons consistently responsive to the target concept are selected. Finally, FIA constructs masks from the identified neurons and fuses them into a unified multi-concept mask, where Concept Agnostic Neurons that broadly support general content generation are preserved while concept-specific neurons are pruned to remove the targets. FIA is training-free and requires minimal hyperparameter tuning for new tasks, enabling plug-and-play use. Extensive experiments across three distinct unlearning tasks demonstrate that FIA achieves more reliable multi-concept unlearning, improving forgetting effectiveness while maintaining generation fidelity and quality. Code is available at https://github.com/kaiyuan02415/Forget-It-All

---

### A universal vision transformer for fast calorimeter simulations

Authors: Luigi Favaro, Andrea Giammanco, Claudius Krause
Published: 2026-01-07
Categories: hep-ph, cs.LG, hep-ex, physics.ins-det
arXiv: https://arxiv.org/abs/2601.05289v2

The high-dimensional complex nature of detectors makes fast calorimeter simulations a prime application for modern generative machine learning. Vision transformers (ViTs) can emulate the Geant4 response with unmatched accuracy and are not limited to regular geometries. Starting from the CaloDREAM architecture, we demonstrate the robustness and scalability of ViTs on regular and irregular geometries, and multiple detectors. Our results show that ViTs generate electromagnetic and hadronic showers with minimal deviations from Geant4 in multiple evaluation metrics, while maintaining the generation time in the $\mathcal{O}(10-100)$ ms on a single GPU. Furthermore, we show that pretraining on a large dataset and fine-tuning on the target geometry leads to reduced training costs and higher data efficiency, or altogether improves the fidelity of generated showers.

---

### UniDrive-WM: Unified Understanding, Planning and Generation World Model for Autonomous Driving

Authors: Zhexiao Xiong, Xin Ye, Burhan Yaman, Sheng Cheng, Yiren Lu, Jingru Luo, Nathan Jacobs, Liu Ren
Published: 2026-01-07
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.04453v4

World models have become central to autonomous driving, where accurate scene understanding and future prediction are crucial for safe control. Recent work has explored using vision-language models (VLMs) for planning, yet existing approaches typically treat perception, prediction, and planning as separate modules. We propose UniDrive-WM, a unified VLM-based world model that jointly performs driving-scene understanding, trajectory planning, and trajectory-conditioned future image generation within a single architecture. UniDrive-WM's trajectory planner predicts a future trajectory, which conditions a VLM-based image generator to produce plausible future frames. These predictions provide additional supervisory signals that enhance scene understanding and iteratively refine trajectory generation. We further compare discrete and continuous output representations for future image prediction, analyzing their influence on downstream driving performance. Experiments on the challenging Bench2Drive benchmark show that UniDrive-WM produces high-fidelity future images and improves planning performance by 7.3% in L2 trajectory error and 10.4% in collision rate over the previous best method. These results demonstrate the advantages of tightly integrating VLM-driven reasoning, planning, and generative world modeling for autonomous driving. The project page is available at https://unidrive-wm.github.io/UniDrive-WM.

---

### Explainable Admission-Level Predictive Modeling for Prolonged Hospital Stay in Elderly Populations: Challenges in Low- and Middle-Income Countries

Authors: Daniel Sierra-Botero, Ana Molina-Taborda, Leonardo Espinosa-Leal, Alexander Karpenko, Alejandro Hernandez, Olga Lopez-Acevedo
Published: 2026-01-07
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.04449v1

Prolonged length of stay (pLoS) is a significant factor associated with the risk of adverse in-hospital events. We develop and explain a predictive model for pLos using admission-level patient and hospital administrative data. The approach includes a feature selection method by selecting non-correlated features with the highest information value. The method uses features weights of evidence to select a representative within cliques from graph theory. The prognosis study analyzed the records from 120,354 hospital admissions at the Hospital Alma Mater de Antioquia between January 2017 and March 2022. After a cleaning process the dataset was split into training (67%), test (22%), and validation (11%) cohorts. A logistic regression model was trained to predict the pLoS in two classes: less than or greater than 7 days. The performance of the model was evaluated using accuracy, precision, sensitivity, specificity, and AUC-ROC metrics. The feature selection method returns nine interpretable variables, enhancing the models' transparency. In the validation cohort, the pLoS model achieved a specificity of 0.83 (95% CI, 0.82-0.84), sensitivity of 0.64 (95% CI, 0.62-0.65), accuracy of 0.76 (95% CI, 0.76-0.77), precision of 0.67 (95% CI, 0.66-0.69), and AUC-ROC of 0.82 (95% CI, 0.81-0.83). The model exhibits strong predictive performance and offers insights into the factors that influence prolonged hospital stays. This makes it a valuable tool for hospital management and for developing future intervention studies aimed at reducing pLoS.

---