Machine Learning · Cross-listed from cs.CL
PSK@EEUCA 2026: Fine-Tuning Large Language Models with Synthetic Data Augmentation for Multi-Class Toxicity Detection in Gaming Chat
Srikar Kashyap Pulipaka
This paper describes our system for the EEUCA 2026 Shared Task on Understanding Toxic Behavior in Gaming Communities. The task involves classifying World of Tanks chat messages into six toxicity categories: Non-toxic, Insults/Flaming, Other Offensive, Hate/Harassment, Threats, and Extremism. We explore multiple approaches including encoder-based models, instruction-tuned LLMs with LoRA fine-tuning, hierarchical classification, one-vs-rest strategies, and various ensemble methods. Our best system combines Llama 3.1 8B with carefully calibrated 5% synthetic data augmentation, achieving an F1-macro score of 0.6234 on the test set, placing 4th out of 35 participating teams. We provide extensive analysis of the
dataset's annotation patterns and their impact on model generalization, revealing a critical ”validation trap” phenomenon where high validation performance correlates with poor test transfer.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.AI
Three-in-One World Model: Energy-Based Consistency, Prediction, and Counterfactual Inference for Marketing Intervention
Junichiro Niimi
Marketing decisions reflect the interaction of latent consumer heterogeneity, time-varying internal states, and explicit interventions, a structure that current prediction- and
language-oriented models do not capture in a unified manner. We propose a Three-in-One world-model architecture in which a Deep Boltzmann Machine (DBM) learns a frozen belief representation from demographics, time, and lagged actions and outcomes, with lightweight task-specific adapters attached on top. The same belief supports three tasks within a single framework: (i) energy-based consistency evaluation through the DBM's free energy, (ii) outcome prediction through adapters, and (iii) counterfactual inference by holding the belief fixed and varying only the action input given to the adapter. Using a controlled simulation in which the latent price sensitivity, promotion responsiveness, and base preference of each consumer are known, we show that the adapters match a strong MLP baseline on visit- and purchase-AUC while recovering heterogeneous treatment effects substantially better than S-, T-, X-, and DR-learner meta-learners and a Causal Forest baseline built on the same raw features, with the largest gap on a confounded price-promotion intervention. Complementing this, free-energy clamps systematically penalize counterfactual purchase trajectories that lack prior promotional exposure, and the penalty itself depends on the latent base preference in the expected direction. These results indicate that DBM beliefs disentangle latent traits in a form that survives counterfactual queries, providing an integrated world-model substrate for marketing intervention.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CV
Bincheng Peng, Guang Li, Ping Liu et al. (+2 more)
Dataset distillation compresses a large training set into a small synthetic set that preserves downstream training utility. While most existing methods target training networks from scratch, modern visual transfer learning often uses frozen pre-trained encoders followed by lightweight linear probing. Existing
distillation methods for this setting either unroll iterative linear-probe updates with trajectory-based gradient matching, or rely on closed-form formulations originally designed for from-scratch training with neural-tangent-kernel (NTK) approximations. Neither route exploits the fact that frozen-feature linear probing admits a closed-form solution determined directly by the pre-trained features themselves, with no infinite-width approximation and no inner-loop trajectory. We propose Closed-Form Linear-Probe
Dataset Distillation (CLP-DD), a bilevel formulation that computes the linear probe induced by the synthetic set with a sample-space kernel ridge solver. The synthetic images are then updated by evaluating this induced classifier on real features through a temperature-scaled softmax cross-
entropy, where the classifier columns act as learned class anchors in feature space. We further show that the choice of outer objective is decisive: pairing the closed-form inner solver with a standard MSE outer loss substantially underperforms trajectory-based methods, while the discriminative outer loss closes most of the gap. On ImageNet-100 with four pre-trained backbones, CLP-DD substantially improves over LGM without DSA and approaches LGM with DSA at a fraction of the computational cost. On ImageNet-1K, CLP-DD matches or surpasses LGM with DSA on three of four backbones while running roughly $14\times$ faster and using less than one-eighth of the GPU
memory.
Read the full abstract ▾
All authors Bincheng Peng, Guang Li, Ping Liu, Takahiro Ogawa, Miki Haseyama
Machine Learning
Coupling Models for One-Step Discrete Generation
Fred Zhangzhi Peng, Avishek Joey Bose, Anru R. Zhang et al. (+1 more)
Generative modeling over discrete structures underpins applications across deep learning, from biological sequence design and code generation to large
language models, yet generation often remains sequential, relying on autoregressive decoding or iterative refinement. In this work, we introduce Coupling Models(Coupling Models), a one-step discrete generative model that learns a direct coupling between discrete sequences and Gaussian latents. Unlike recent
distillation methods that compress a pretrained multi-step sampler into a few steps, Coupling Model trains a purpose-built decoder to invert this coupling and generate samples in a single step. The model also avoids complex continuous flows over the simplex and hand-specified data-to-noise couplings. Empirically,Coupling Model improves the strongest one-step baselines in each domain: it reduces LM1B text-generation perplexity by 33% at its lowest-perplexity operating point, Fly
Brain enhancer-design FBD by 18%, and MNIST-Binary FID by 46%. These results suggest that effective one-step discrete generation depends strongly on how data and noise are coupled before decoding. Code is available at https://github.com/pengzhangzhi/Coupling-Models.
Read the full abstract ▾
All authors Fred Zhangzhi Peng, Avishek Joey Bose, Anru R. Zhang, Alexander Tong
Machine Learning · Cross-listed from cs.CV
Huaiyuan Qin, Muli Yang, Gabriel James Goenawan et al. (+4 more)
A recent work shows that
Attention Transfer, which transfers only the
attention patterns from a pre-trained teacher
Vision Transformer (ViT) to a randomly initialized standard student ViT, is sufficient to recover the full benefit of the teacher's pre-trained weights. We revisit this finding on a comprehensive
benchmark of 20 teachers from 11 well-known ViT families and reveal that
Attention Transfer is not universally effective. While 7 families transfer successfully, 4 consistently fail, falling up to 5.1% below the from-scratch no-transfer baseline. Further results demonstrate that this failure is family-consistent across model sizes, and persists under extended training durations, different transfer
datasets, and out-of-distribution evaluations. Controlled analyses then consistently localize the problem to the
attention-routing channel, indicating that the key issue is not whether the student can match the teacher's
attention patterns, but whether the matched patterns remain functional for the student. Crucially, we identify architectural mismatch between the pre-trained teacher and the standard student as the primary mechanism. By adding only the teacher's native architectural components to the student in a randomly initialized state, we completely reverse the failure for all 4 families. Notably, these components alone do not improve from-scratch training, confirming that they specifically unlock the usability of the teacher's
attention. We further show that this failure is not explained by the inadequate choice of transfer loss, differences in pre-training recipes, or the intrinsic quality of
attention maps. Our findings refine the prevailing understanding of
attention in ViT representations:
attention is sufficient only when the student architecture matches the teacher.
Read the full abstract ▾
All authors Huaiyuan Qin, Muli Yang, Gabriel James Goenawan, Peng Hu, Chen Gong, Xi Peng, Hongyuan Zhu
Machine Learning
Star Elastic: Many-in-One Reasoning LLMs with Efficient Budget Control
Ali Taghibakhshi, Ruisi Cai, Saurav Muralidharan et al. (+17 more)
Training a family of large
language models (LLMs), either from scratch or via iterative compression, is prohibitively expensive and inefficient, requiring separate training runs for each model in the family. In this paper, we introduce
Star Elastic, a novel LLM post-training method that adds N nested submodels to a given parent
reasoning model using the compute of one run (N-fold savings) via a single post-training job. Beyond reducing training costs,
Star Elastic also addresses a fundamental limitation of efficient
reasoning: the
rigidity of static architectures, which forces the allocation of constant resources regardless of
token difficulty. By unlocking elastic budget control,
Star Elastic enables a novel inference scheme that uses different submodels for each
reasoning phase (thinking and answering).
Star Elastic supports (1) nesting along the SSM,
embedding channel, MoE, and FFN axes, (2) learning nested submodels via an end-to-end trainable router, and (3) curriculum-based knowledge
distillation. Building on the Nemotron Elastic framework, we apply
Star Elastic to the NVIDIA Nemotron Nano models, with a particular focus on hybrid Mixture-of-Experts (MoE) architectures: from Nemotron Nano v3 (30B/3.6A), we generate 23B (2.8A) and 12B (2.0A) variants with 160B training
tokens. All nested models match or outperform independently trained baselines of comparable size and achieve a 360x reduction versus pretraining from scratch and a 7x reduction over state-of-the-art compression. Crucially, elastic budget control advances the accuracy-latency Pareto frontier, achieving up to 16% higher accuracy and 1.9x lower latency via dynamic per-phase model selection. We further extend
Star Elastic to quantized regimes via Quantization-Aware
Distillation (QAD), producing nested NVFP4 and FP8 elastic checkpoints that preserve zero-shot slicing while delivering smaller deployment footprints.
Read the full abstract ▾
All authors Ali Taghibakhshi, Ruisi Cai, Saurav Muralidharan, Sharath Turuvekere Sreenivas, Aditya Vavre, Ameya Sunil Mahabaleshwarkar, Bilal Kartal, Sheldon Liang, Marcin Chochowski, Zijia Chen, Akhiad Bercovich, Ran Zilberstein, Ran El-Yaniv, Yonatan Geifman, Daniel Korzekwa, Yoshi Suhara, Oluwatobi Olabiyi, Ashwath Aithal, Nima Tajbakhsh, Pavlo Molchanov
Machine Learning
HyperEyes: Dual-Grained Efficiency-Aware Reinforcement Learning for Parallel Multimodal Search Agents
Guankai Li, Jiabin Chen, Yi Xu et al. (+2 more)
Existing
multimodal search
agents process target entities sequentially, issuing one tool call per entity and accumulating redundant interaction rounds whenever a query decomposes into independent sub-
retrievals. We argue that effective
multimodal agents should search wider rather than longer: dispatching multiple grounded queries concurrently within a round. To this end, we present HyperEyes, a parallel
multimodal search
agent that fuses visual grounding and
retrieval into a single atomic action, enabling concurrent search across multiple entities while treating inference efficiency as a first-class training objective. HyperEyes is trained in two stages. For cold-start supervision, we develop a Parallel-Amenable Data Synthesis Pipeline covering visual multi-entity and textual multi-constraint queries, curating efficiency-oriented trajectories via Progressive Rejection Sampling. Building on this, our central contribution, a Dual-Grained Efficiency-Aware Reinforcement Learning framework, operates at two levels. At the macro level, we propose TRACE (Tool-use Reference-Adaptive Cost Efficiency), a trajectory-level reward whose reference is monotonically tightened during training to suppress superfluous tool calls without restricting genuine multi-hop search. At the micro level, we adapt On-Policy
Distillation to inject dense
token-level corrective signals from an external teacher on failed rollouts, mitigating the credit-assignment deficiency of sparse outcome rewards. Since existing
benchmarks evaluate accuracy as the sole metric, omitting inference cost, we introduce IMEB, a human-curated
benchmark of 300 instances that jointly evaluates search capability and efficiency. Across six
benchmarks, HyperEyes-30B surpasses the strongest comparable open-source
agent by 9.9% in accuracy with 5.3x fewer tool-call rounds on average.
Read the full abstract ▾
All authors Guankai Li, Jiabin Chen, Yi Xu, Xichen Zhang, Yuan Lu
Machine Learning
Learning Multi-Relational Graph Representations for DNA Methylation-Based Biological Age Estimation
Qing Qing, Xikun Zhang, Zhongyuan Zhang et al. (+7 more)
Aging clocks aim to estimate biological age, a measure of physiological state distinct from chronological age, from observable biomarkers, and are widely used for health assessment and disease analysis. DNA methylation is a particularly informative biomarker due to its stability and strong association with aging, and recent learning-based approaches have improved predictive performance. However, most existing methods treat CpG sites as independent features, overlooking the complex and heterogeneous biological relationships among them. We propose RelAge-GNN, a multi-relational graph neural network framework for DNA methylation-based age prediction. Our method constructs three complementary graphs capturing co-methylation patterns, genomic co-localization, and gene-level associations among CpG sites. Each graph is modeled by an independent GNN branch, and a learnable gating mechanism adaptively fuses the resulting representations. Experiments on large-scale
datasets show that RelAge-GNN achieves competitive accuracy and stronger correlation with chronological age compared to state-of-the-art methods. Moreover, the model exhibits improved sensitivity in detecting age acceleration across diverse disease cohorts, highlighting its potential utility for disease characterization. Finally, through post hoc interpretability analyses, we quantify the contributions of different relational structures and CpG sites, providing biologically meaningful insights and suggesting potential directions for aging-related research. Our code is available at: https://anonymous.4open.science/r/RelAge-GNN-F1E3/.
Read the full abstract ▾
All authors Qing Qing, Xikun Zhang, Zhongyuan Zhang, Jiarui Liu, Xingtong Yu, Xiaotao Shen, Ziqi Xu, Qixin Zhang, Zhe Wang, Renqiang Luo
Machine Learning
Cost-Ordered Feasibility for Multi-Armed Bandits with Cost Subsidy
Ishank Juneja, Carlee Joe-Wong and Osman Yağan
The classic multi-armed bandit (MAB) problem tackles the challenge of accruing maximum reward while making decisions under uncertainty. However, in applications, often the goal is to minimize cost subject to a constraint on the minimum permissible reward, an objective captured by multi-armed bandits with cost-subsidy (MAB-CS). Of interest to this paper is the setting where the quality (reward) constraint is specified relative to the unknown best reward and the cost of each arm is known. We characterize the expected sub-optimal samples required by any policy by proving instance-dependent lower bounds that offer new insight into the problem and are a strict generalization of prior bounds. Then, we propose an
algorithm called Cost-Ordered Feasibility (COF) that leverages our insight and intelligently combine samples from all arms to gauge the feasibility of a cheap arm. Thereafter, we analyze COF to establish instance-dependent upper bounds on its expected cumulative cost and quality regret, i.e., relative to the cheapest feasible arm. Finally, we empirically validate the merits of COF, comparing it to baselines from the literature through extensive simulation experiments on the MovieLens and Goodreads
datasets as well as representative synthetic instances. Not only does our paper develop qualitatively better theoretical regret upper bounds, but COF also convincingly demonstrates improved empirical performance.
Read the full abstract ▾
Machine Learning
Neurosymbolic Imitation Learning with Human Guidance: A Privileged Information Approach
Nikhilesh Prabhakar, Varun Balaji, Athresh Karanam et al. (+2 more)
Imitation learning is widely used for learning to act in complex environments. While pure neural-based methods handle high dimensional data effectively, they suffer from the requirement of large number of samples and are prone to overfitting. Pure symbolic approaches, while generalize well, do not handle high-dimensional data effectively. We propose a neurosymbolic approach that achieves the best of both worlds, i.e, handling high-dimensional data while achieving generalization. The key advantage of our approach is that it can effectively exploit additional privileged information that is available only during training (in our case, gaze data). Our empirical evaluations demonstrate the effectiveness, efficiency and the generalization capability of our proposed approach.
Read the full abstract ▾
All authors Nikhilesh Prabhakar, Varun Balaji, Athresh Karanam, Kristian Kersting, Sriraam Natarajan
Machine Learning · Cross-listed from cs.IR
Topic Is Not Agenda: A Citation-Community Audit of Text Embeddings
Junseon Yoo
Vector search and
retrieval-augmented generation (RAG) rest on the assumption that cosine similarity between text
embeddings reflects conceptual relatedness. We measure where this assumption breaks. We build an augmented citation graph over 3.58M scientific papers and partition it via Leiden CPM at two granularities: sub-field (L1) and research-agenda (L2, hierarchical inside each L1). Four state-of-the-art
embeddings (Gemini, Qwen3-8B, Qwen3-0.6B, SPECTER2) clear the L1 bar reasonably (45-52% top-10 same-rate) but stop working at L2: only 15-21% of top-10 neighbors share the query's research agenda. In absolute terms, 8 of every 10 retrieved papers are off-agenda. The failure is universal across eight scientific domains and all four models; SPECTER2, despite its citation-based contrastive training, is the weakest. As a diagnostic probe, we test whether the same augmented graph also functions as a
retrieval signal: a deliberately simple citation-count rerank reaches 57.7% top-1 L2 on top of LLM-expanded Boolean
retrieval and 59.6% on top of plain BM25, on 80 curated agenda queries – about 9 points above the best cosine retriever (Gemini, 50.6%) and 20 points above BM25 alone (39.3%). The probe isolates a slice of the agenda-matching signal the graph carries but the
embeddings miss, connecting recent theoretical limits on single-vector
retrieval to a concrete failure mode of scientific RAG.
Read the full abstract ▾
Machine Learning
Learned Lagrangian Models of PDEs via Euler-Lagrange Residual Minimization
Lyra Zhornyak, Eric Forgoston and M. Ani Hsieh
We present the first method to directly use a learned continuous Lagrangian to forecast the dynamics of systems governed by partial differential equations, exploiting the inherent conservative structure to achieve stable long-range predictions. We develop an optimization-based integrator that minimizes the squared Euler–Lagrange residual via a mesh-free near-
symplectic construction on local space-time patches. Different from integrators for analytical models, integrators for learned models should decouple model error (phase error) from integration error (conservation error). By relying on optimization rather than time-stepping, we bypass the global coupling inherent to fixed discretizations, which slows time- and space-stepping and complicates learning. Our method scales linearly with domain size via Jacobi iteration, and places no structural requirements on the learned network, allowing it to be coupled with existing physics-guided machine learning (ML) methods. We validate our approach on a learned representation of a double pendulum, a one-dimensional wave equation, and a two-dimensional wave equation. Our method achieves error comparable to classical
symplectic methods while generalizing to spatially varying dynamics and arbitrary boundary conditions without retraining.
Read the full abstract ▾
Machine Learning
Regret-Oracle Complexity Tradeoffs in Agnostic Online Learning
Idan Attias, Steve Hanneke and Arvind Ramaswami
Agnostic online learning is classically solved via a reduction to the realizable setting, utilizing Littlestone's Standard Optimal
Algorithm (SOA) as a base learner. However, the SOA is computationally intractable to execute even for a single round. To overcome this barrier, recent work in oracle-efficient online learning replaces the SOA with a realizable base learner that accesses the concept class exclusively through an offline empirical risk minimization (ERM) oracle. While such agnostic learners achieve near-optimal expected regret, they suffer from a doubly-exponential oracle complexity of $O\big(T^{2^{O(d_\mathrm{LD})}}\big)$, where $d_\mathrm{LD}$ is the Littlestone dimension and $T$ is the number of rounds. In this work, we significantly improve this oracle complexity while relying on an even weaker primitive: a weak-consistency oracle, which merely decides whether a given labeled
dataset is realizable. At the core of our approach is an adaptive and dynamic agnostic-to-realizable reduction that actively prunes non-realizable label sequences on the fly. By using the VC dimension ($d_\mathrm{VC}$) to bound the number of dynamically maintained active paths, our
algorithm reduces the total query complexity down to $O(T^{d_\mathrm{VC}+1})$ while perfectly preserving near-optimal expected regret. Crucially, this dynamic
pruning also yields a
memory reduction over the standard reduction. Furthermore, we formally quantify the regret–oracle complexity tradeoff, providing upper bounds that smoothly interpolate between restricted query budgets and attainable expected regret. We complement these with lower bounds proving that any learner restricted to $Q = o(\sqrt{T})$ queries must suffer an expected regret of $\Omega(T/Q)$.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.LO
MathlibPR: Pull Request Merge-Outcome Benchmark for Formal Mathematical Libraries
Zixuan Xie, Xinyu Liu and Shangtong Zhang
The ecosystem of Lean and Mathlib has become the de facto standard for large
language model (LLM) assisted formal
reasoning with remarkable successes in recent years. Those successes, however, only consume Mathlib as an essential dependency but do not directly contribute to it. In the meantime, the growth of Mathlib has recently been bottlenecked by the review process, which requires human reviewers to judge whether proposed pull requests (PRs) follow the Mathlib's conventions and are worth integrating as part of a shared mathematical infrastructure. This leads to our central question: can LLMs help review Mathlib PRs? To this end, we introduce MathlibPR, a
benchmark built from real Mathlib4 PR histories. We further propose a staged evaluation protocol and use it to evaluate both LLM models (e.g., DeepSeek, Qwen, Goedel, and Kimina) and LLM
agents (e.g., Codex and Claude Code). Surprisingly, both LLM models and LLM
agents struggle to distinguish merge-ready PRs from build-passing PRs that were revised or never merged. By turning Mathlib PR histories into a supervised signal, MathlibPR provides a step toward reviewer assistants and reward models that could help evaluate PRs and steer LLMs toward producing merge-ready Mathlib contributions.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CL
Structural Rationale Distillation via Reasoning Space Compression
Jialin Yang, Jiankun Wang, Jiajun Wu et al. (+3 more)
When distilling
reasoning from large
language models (LLMs) into smaller ones, teacher rationales for similar problems often vary wildly in structure and strategy. Like a chef who makes the same dish differently each time, this inconsistency burdens the student with noisy supervision that is hard to internalize. We propose
Distillation through
Reasoning Path Compression (D-RPC), which constrains the teacher to follow a compact, dynamically maintained bank of reusable high-level
reasoning paths. For each training question, D-RPC retrieves the most relevant path and conditions the teacher to follow it, producing rationales that are consistent across similar problems yet diverse enough to cover different problem types. A PAC-Bayes analysis formalizes a trade-off between bank size and coverage, with smaller banks limiting path diversity at the risk of coverage gaps. Our ablations favor an intermediate bank size, consistent with this trade-off. Across five math and commonsense
reasoning benchmarks with two student models, D-RPC achieves the highest average accuracy among the evaluated
distillation methods, while using fewer
tokens than template-heavy alternatives.
Read the full abstract ▾
All authors Jialin Yang, Jiankun Wang, Jiajun Wu, Henry Leung, Jiayu Zhou, Steve Drew
Machine Learning · Cross-listed from cs.AI
Can You Break RLVER? Probing Adversarial Robustness of RL-Trained Empathetic Agents
Deeraj S K, Sadhana Devarajan, Krishna Mehra et al. (+1 more)
Reinforcement learning from verifiable
emotion rewards (RLVER) yields strong empathetic
language models, but is evaluated with cooperative, honest simulated users. Real users gaslight, escalate, and demand unconditional validation. We introduce the
Adversarial Empathy
Benchmark (AEB), six
adversarial trajectory types whose discriminative rewards penalize formulaic comfort, and the Emotional Consistency Score (ECS), which separates whether a model improves the user's emotional state from whether that state stays legible in the transcript. Across eight scenario-matched conditions (480 dialogues; two RLVER and two Qwen2.5 base models, with and without thinking), RLVER-PPO-Think outperforms its untuned same-scale baseline (FS $0.979$ vs.\ $0.761$, $p<10^{-12}$) with no dialogue collapses and 53% higher hidden-intention detection; the gain is not explained by response length. Yet under an independent cross-family judge (Gemma-3-4b-it), ECS is indistinguishable between the two ($p=0.43$). We call this an outcome–legibility dissociation: RLVER improves emotional outcomes without making the user's state more legible. We do not claim ECS measures internal understanding.
Read the full abstract ▾
All authors Deeraj S K, Sadhana Devarajan, Krishna Mehra, Sudhakar Mishra
Machine Learning
Adaptive Negative Reinforcement for LLM Reasoning:Dynamically Balancing Correction and Diversity in RLVR
Yash Ingle, Jaival Chauhan, Ankit Yadav et al. (+1 more)
Reinforcement learning with verifiable rewards (RLVR) has become a highly effective method for improving the
reasoning abilities of Large
Language Models (LLMs). Recent research shows that Negative Sample Reinforcement (NSR) – which focuses on penalizing incorrect steps rather than simply rewarding correct ones – can match or even exceed the performance of more complex frameworks like PPO and GRPO across the entire Pass@k spectrum. However, current NSR techniques usually apply a fixed penalty throughout the training process and treat every incorrect response with the same weight. To address these limitations, we propose two extensions to the NSR framework: Adaptive Negative Sample Reinforcement. Rather than using a fixed update rule, A-NSR uses time-dependent scheduling functions. In the initial training phases, the system focuses heavily on correcting errors to stabilize the model. As training continues, it shifts toward more subtle and controlled updates. We also introduce Confidence-Weighted Negative Reinforcement, which operates on the principle that different mistakes carry different levels of importance. CW-NSR assigns specific penalty weights based on the model's normalized sequence likelihood. If the model is highly confident in a wrong path, it receives a larger penalty and for uncertain errors – where the model is effectively exploring – are penalized less strictly. Our formal analysis shows how these mechanisms govern
token-level updates, allowing the model to leverage prior-guided probability redistribution while providing a natural defense against overfitting. We evaluated these methods on difficult
reasoning datasets, including MATH, AIME 2025, and AMC23, using the Qwen2.5-Math-1.5B architecture.
Read the full abstract ▾
All authors Yash Ingle, Jaival Chauhan, Ankit Yadav, Sudhakar Mishra
Machine Learning
GAD in the Wild: Benchmarking Graph Anomaly Detection under Realistic Deployment Challenges
Jingjing Zhou, Shiyu Huang, Qing Qing et al. (+7 more)
Graph Anomaly Detection (GAD) is a critical task in graph machine learning with vital applications in financial fraud detection and social platform governance. However, existing GAD
benchmarks are often restricted to small-scale, curated graphs with relatively balanced anomaly ratios, leaving a substantial gap between academic evaluation and real-world deployment. To bridge this gap, we present a multi-dimensional
benchmark that systematically evaluates GAD models under three deployment-relevant challenges: million-scale graphs, extreme anomaly scarcity, and missing node attributes. We derive a family of controlled
benchmark variants from five diverse graphs, including two native industrial-scale
datasets with over 3.7 million nodes. Our extensive evaluation of nine representative GAD models reveals three major limitations: (1) most GNN-based methods fail to scale to million-node graphs due to prohibitive
memory requirements; (2) detection performance drops sharply under realistic anomaly ratios (e.g., 0.1%), often resulting in zero recall; and (3) reconstruction-based models are highly sensitive to attribute imputation strategies. Our findings suggest that strong performance in laboratory settings does not guarantee robustness in production environments. We release this
benchmark and empirical evaluation as a diagnostic testbed to promote the development of robust and scalable GAD systems for large-scale, imperfect graphs encountered in practice. Code is available at https://anonymous.4open.science/r/
Benchmark_GAD-E7A3.
Read the full abstract ▾
All authors Jingjing Zhou, Shiyu Huang, Qing Qing, Zuquan Yuan, Huafei Huang, Ziqi Xu, Mingliang Hou, Xikun Zhang, Renqiang Luo, Ivan Lee
Machine Learning
Simple KNN-Based Outlier Detection Achieves Robust Clustering
Tianle Jiang and Yufa Zhou
Being robust to the presence of outliers is crucial for applying clustering
algorithms in practice. In the $\textit{robust $k$-Means}$ problem (i.e., $k$-Means with outliers), the goal is to remove $z$ outliers and minimize the $k$-Means cost on the remaining points. Despite the close connection between robust $k$-Means and outlier detection, both theoretical and empirical understanding of the effectiveness of $\textit{classic outlier detection heuristics}$ for robust $k$-Means remains limited. In this paper, we prove that under a practical assumption on the optimal cluster sizes, simply removing points with large $K$-Nearest-Neighbor distances achieves performance comparable to prior work in terms of approximation guarantees: it yields a constant-factor reduction from robust $k$-Means to standard $k$-Means, without introducing additional centers or discarding extra outliers, as is commonly required by existing approaches. Empirically, experiments on real-world
datasets show that our method outperforms or matches several more sophisticated
algorithms in terms of clustering cost and runtime. These results demonstrate that simple KNN-based heuristics can be surprisingly effective for robust clustering, highlighting new opportunities to bridge techniques from outlier detection and clustering.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.IR
RRCM: Ranking-Driven Retrieval over Collaborative and Meta Memories for LLM Recommendation
Shijun Li, Pranav Belligundu, Tianxin Wei et al. (+3 more)
Large
Language Models (LLMs) have emerged as a promising paradigm for next-generation recommender systems, offering strong semantic understanding and natural-
language reasoning abilities. Despite recent progress, current LLM-based recommenders still face key challenges in constructing decision-relevant contexts from heterogeneous evidence. First, existing methods often rely on fixed context construction strategies: collaborative behavioral evidence and item-side metadata are typically incorporated through predefined
prompts, static
retrieval pipelines, or handcrafted injection mechanisms, making it difficult to determine what information is truly beneficial for each instance. Second, heterogeneous evidence introduces a severe context-efficiency bottleneck. Rich metadata and collaborative interaction records can quickly overwhelm the context window, while aggressive compression or heuristic filtering may discard fine-grained evidence critical for accurate recommendation. To address these challenges, we propose RRCM, a ranking-driven
retrieval-and-
reasoning framework over collaborative and metadata
memories for LLM-based agentic recommendation. RRCM starts from a lightweight user-history context and learns whether to recommend directly, retrieve collaborative evidence, retrieve item metadata, or interleave both through
reasoning. Both
memories are represented in natural
language and accessed through a unified
retrieval interface, enabling flexible evidence acquisition without handcrafted CF injection or fixed
retrieval rules. We optimize this
memory-reading policy with an outcome-only ranking reward, instantiated using group relative policy optimization, so that
retrieval decisions are directly driven by final top-k recommendation quality. Extensive experiments show that RRCM significantly outperforms traditional baselines and diverse LLM-based recommendation approaches.
Read the full abstract ▾
All authors Shijun Li, Pranav Belligundu, Tianxin Wei, Wooseong Yang, Yu Wang, Joydeep Ghosh
Machine Learning
The Position Curse: LLMs Struggle to Locate the Last Few Items in a List
Zhanqi Zhang, Hua-Dong Xiong, Robert C. Wilson et al. (+3 more)
Modern large
language models (LLMs) can find a needle in a haystack (locating a single relevant fact buried among hundreds of thousands of irrelevant
tokens) with near-saturated accuracy, yet fail to retrieve the last few items in a short list. We call this failure the Position Curse. For instance, even in a two-line code snippet, Claude Opus 4.6 misidentifies the second-to-last line most of the time. To characterize this failure, we evaluated two complementary queries: given a position in a sequence (of letters or words), retrieve the corresponding item; and given an item, return its position. Each position is specified as a forward or backward offset from an anchor, either an endpoint of the list (its start or end) or another item in the list. Across both open-source and frontier closed-source models, backward
retrieval substantially lags forward
retrieval. To test whether this capability can be rescued by post-training, we constructed PosBench, a position-focused training
dataset. LoRA fine-tuning improves both forward and backward
retrieval and generalizes to a held-out code-understanding
benchmark (PyIndex), yet absolute performance remains far from saturated. As LLM coding
agents increasingly operate over large codebases where precise indexing becomes essential for code understanding and editing, position-based
retrieval emerges as a key capability for future pretraining objectives and model design.
Read the full abstract ▾
All authors Zhanqi Zhang, Hua-Dong Xiong, Robert C. Wilson, Mikio Aoi, Marcelo G. Mattar, Li Ji-An
Machine Learning
Convergence and Emergence of In-Context Reinforcement Learning with Chain of Thought
Zixuan Xie, Xinyu Liu, Rohan Chandra et al. (+1 more)
In-context reinforcement learning (ICRL) refers to the ability of RL
agents to adapt to new tasks at inference time without parameter updates by conditioning on additional context. Recent empirical studies further demonstrate that Chain-of-Thought (CoT) generation can amplify this ICRL capability. This paper is the first to provide a theoretical understanding on how CoT interacts with ICRL. We conduct our analysis in a policy evaluation setup with linear
Transformer. We prove that with specific
Transformer parameters, the CoT generation process is equivalent to repeatedly executing temporal difference learning updates. Additionally, we provide finite sample convergence analysis showing that the policy evaluation error decreases geometrically with CoT length and eventually saturates at a statistical floor determined by the context length. We also prove that the desired
Transformer parameters are a global minimizer of the pretraining loss, providing a theoretical understanding on the empirical emergence of those parameters.
Read the full abstract ▾
All authors Zixuan Xie, Xinyu Liu, Rohan Chandra, Shangtong Zhang
Machine Learning · Cross-listed from cs.AI
AdaTKG: Adaptive Memory for Temporal Knowledge Graph Reasoning
Seunghan Lee, Jun Seo, Jaehoon Lee et al. (+7 more)
Temporal knowledge graphs (TKGs) represent time-stamped relational facts and support a wide range of
reasoning tasks over evolving events. However, existing methods produce entity representations that are static at the entity level, in that each representation is a function of learned parameters only and retains no trace of the interactions in which the entity has participated. In this paper, we depart from this static view and propose that each entity be modeled as an adaptive process whose representation is refined every time the entity participates in a fact. To this end, we propose AdaTKG, which maintains a per-entity
memory that is updated with every observed interaction, with the
memory accumulating online and predictions improving as more interactions arrive. Specifically, we instantiate the
memory update as a learnable exponential moving average governed by a single shared scalar instead of using learnable parameters for each entity, enabling AdaTKG to handle entities unseen during training. Extensive experiments confirm consistent gains over TKG baselines, demonstrating the effectiveness of adaptive
memory. Code is available at: https://github.com/seunghan96/AdaTKG
Read the full abstract ▾
All authors Seunghan Lee, Jun Seo, Jaehoon Lee, Sungdong Yoo, Minjae Kim, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, SoonYoung Lee, Wonbin Ahn
Machine Learning
When Symbol Names Should Not Matter: A Logistic Theory of Fresh-Symbol Classification
Wenjie Guan and Jelena Bradic
Template tasks have emerged as a clean testbed for asking whether
transformers reason with abstract symbols rather than concrete
token names. We study the fixed-label classification version of this problem, where train and test examples share latent templates but may use disjoint vocabularies. Unlike next-
token prediction, the model need not emit unseen symbols; it must learn a decision rule invariant to symbol renaming. We analyze regularized kernel logistic classification in the
transformer-kernel regime. Our main result decomposes the learned predictor into an ideal template-level classifier and a finite-sample perturbation caused by accidental
token overlaps in the training data. We encode these overlaps by a colored collision graph and prove high-probability margin-transfer guarantees for fresh-symbol classification. This perspective extends template-based analyses to logistic classification and refines scalar diversity conditions: vocabulary size controls the average rate of collisions, but collision geometry controls whether the ideal classification margin is preserved. More broadly, the same perturbation framework applies to abstraction-augmented inputs, yielding a general margin-versus-collision criterion for identifying when prompting strategies improve fresh-symbol generalization. Synthetic template experiments illustrate the predicted roles of regularization, sample size, and
transformer-kernel structure.
Read the full abstract ▾
Machine Learning · Cross-listed from stat.ML
Classification Fields: Arbitrarily Fine Recursive Hierarchical Clustering From Few Examples
Yicen Li, Ruiyang Hong, Anastasis Kratsios et al. (+2 more)
Classical clustering methods usually return either a finite partition of the observed data or a finite dendrogram over it. This finite-sample view is inadequate when the hierarchy of interest is a recursive
geometric object with fine-scale refinements that continue beyond the levels directly observed. We introduce classification fields: infinite-depth hierarchical cluster structures on $\mathbb{R}^d$ generated by a local parent-to-child refinement rule. A classification field generator maps each parent centre to an ordered, bounded, and separated tuple of child residuals. Together with a root and a scale factor, this rule recursively generates cluster centres, Voronoi
cells, and a metric DAG encoding the hierarchy. Given only a finite prefix of such a hierarchy, we learn a classification field predictor that approximates the generator and can be rolled out to unseen depths. We prove exponential truncation convergence in the completed
cell metric and ReLU realizability with width $O(\varepsilon^{-\gamma})$ and depth $\widetilde O(\varepsilon^{-3\gamma/2})$, where $\gamma=\log K/(-\log s)$, up to finite-window aspect-ratio factors. The approximation holds at the level of the induced compact metric structures, measured in the completed
cell-metric Hausdorff distance. Experimental validation on matched CFG-generated hierarchies, IFS fractals, and image-induced recursive clustering hierarchies shows that learned predictors preserve ordered child slots, unordered geometry, and hierarchy-level path metrics under recursive rollout. These results support the claim that finite hierarchical observations can reveal local refinement rules capable of generating substantially deeper classification fields.
Read the full abstract ▾
All authors Yicen Li, Ruiyang Hong, Anastasis Kratsios, Haitz Sáez de Ocáriz Borde, Paul D. McNicholas
Machine Learning
Monotone Neural Policy Iteration for High-Dimensional First-Order Hamilton–Jacobi–Bellman Equations
Minseok Kim, Yeongjong Kim, Namkyeong Cho et al. (+1 more)
We analyze a neural semi-discrete method for high-dimensional first-order Hamilton-Jacobi-Bellman (HJB) equations with known or learned dynamics. Centered differences and an artificial viscosity $Nh=O(h)$ define a monotone operator evaluated through $2d+1$ shifted network queries; policy iteration solves the resulting Bellman equation without a tensor grid. At fixed $h$, the sharp componentwise condition $\max_i|f_i|\le2N$ turns every frozen-policy operator into a nearest-neighbor Markov-chain generator with a policy-independent total jump rate. Uniformization gives whole-space well-posedness for measurable feedbacks, an explicit Poisson-tail bound on the numerical domain of dependence, and boundary-free localization. The representation also yields a posteriori policy-evaluation bounds that account for residual and learned-model errors. A greedy-gap analysis controls inexact policy iteration at fixed $h$; a separate consistency estimate connects the semi-discrete equation to the continuous HJB equation. Experiments reproduce the extremal tail, show rates consistent with $O(\sqrt h)$ and nearly $h$-independent exact-policy-iteration decay, and assess empirical
estimator effectivity. A nonsmooth example shows that the continuous residual can miss a non-viscosity solution, whereas the shifted residual detects the defect. Further tests provide a structured interval-verified certificate calibration, an early-budget benefit of policy freezing for bang-bang control, and learned-dynamics diagnostics. A structured nonlinear problem with active compact-control constraints is tested against a manufactured semi-discrete reference through $d=1024$.
Read the full abstract ▾
All authors Minseok Kim, Yeongjong Kim, Namkyeong Cho, Yeoneung Kim
Machine Learning
Conformal-Style Quantile Analyses for Stochastic Bandits
Chengyu Du and Mengfan Xu
Stochastic bandit
algorithms are usually analyzed under a mean-reward criterion, yet many problems favor arms with strong upper-tail performance, which we study herein. For a fixed miscoverage level \(\alpha\), the natural upper-tail target of arm \(j\) is the upper endpoint \(F_j^{-1}(1-\alpha/2)\) of a central prediction interval. This target can rank arms differently from their means, creating a central mismatch with the classical bandit objective. To this end, we propose ACP-UCB1, a conformal-style policy that combines an adaptive conformal estimate of the upper endpoint with a UCB-type optimism bonus. The technical challenge is that the conformity scores used by ACP-UCB1 are recomputed from evolving empirical
quantile estimates and evaluated at an adaptive level. We control this endpoint through reward-
quantile concentration, a perturbation argument for recomputed score
quantiles, and deterministic localization of the adaptive level. ACP-UCB1 achieves logarithmic upper-
quantile regret with per-arm contribution \(O(\nicefrac{\log n}{\Delta_j^{\mathrm{ACP}}})\). We also provide metric-specific regret decompositions comparing ACP-UCB1 with UCB1 and use numerical experiments to validate performance and improvement.
Read the full abstract ▾
Machine Learning
Where to Spend Rollouts: Hit-Utility Optimal Rollout Allocation for Group-Based RLVR
Tao Wang, Shuo Li, Yan Sun et al. (+2 more)
Reinforcement learning with verifiable rewards (RLVR) has emerged as a central paradigm for improving the
reasoning capabilities of large
language models. Group-based policy optimization methods, such as GRPO, typically allocate a fixed number of rollouts to every
prompt. This uniform allocation can be inefficient: it over-allocates compute to
prompts whose sampled groups are already saturated while under-exploring
prompts for which additional samples may reveal useful correct trajectories. To address this limitation, we introduce hit utility, the posterior probability that at least one rollout in a proposed additional allocation for a
prompt will be correct. Building on this notion, we propose Hit-Utility Optimal Rollout Allocation (HORA), a learning-free rollout allocation policy that maximizes total posterior hit utility within each allocation batch. HORA adaptively reallocates rollout budgets while leaving the downstream reward evaluation and group-based advantage
estimator unchanged. Across two training sets, four mathematical
reasoning benchmarks, and three model families at multiple scales, HORA preserves comparable Pass@1 and improves Pass@K over compute-matched GRPO, and it outperforms related allocation baselines at large K. It is also drop-in compatible with other group-based
estimators such as RLOO. Ablation studies indicate that the uniform prior used by HORA is competitive with five
prompt-conditioned learned-prior alternatives.
Read the full abstract ▾
All authors Tao Wang, Shuo Li, Yan Sun, Dongsheng Ding, Edgar Dobriban
Machine Learning
Solving Max-Cut to Global Optimality via Feasibility-Preserving Graph Neural Networks
Hao Chen, Chendi Qian, Christopher Morris et al. (+2 more)
Exact solution of hard combinatorial optimization problems often relies on strong convex relaxations, but solving these relaxations repeatedly inside a branch-and-bound
algorithm can be prohibitively expensive. Hence, we consider this challenge for Max-Cut, where branch and bound commonly uses semidefinite programming (SDP) relaxations to bound subproblems. We propose a Max-Cut-specific graph neural network that serves as a principled, lightweight neural proxy for these SDP solvers and can be plugged directly into an exact branch-and-bound framework. The proposed architecture has update steps of complexity $\mathcal{O}(n^2 + ne)$, and predicts both primal- and dual-feasible SDP solutions. The primal SDP solutions yield feasible Max-Cut solutions via the Goemans–Williamson
algorithm. In addition, it is trained in a self-supervised fashion without requiring solved SDP relaxations as labels. Empirically, we show that our architecture can substantially reduce the cost of bounding in exact Max-Cut solving by up to $10.6 \times$ compared with using the state-of-the-art SDP solver Mosek. Our work highlights the potential of learned, validity-preserving surrogates for accelerating exact optimization over structured convex relaxations.
Read the full abstract ▾
All authors Hao Chen, Chendi Qian, Christopher Morris, Andrea Lodi, Can Li
Machine Learning
Lucas Monteiro Paes, Natalie Mackraz, Barry-John Theobald et al. (+1 more)
Language model (LM)
alignment improves model outputs to reflect human preferences while preserving the capabilities of the base model. The most common
alignment approaches are (i) reinforcement learning, which maximizes the expected reward under a KL-divergence constraint, and (ii) best-of-$N$
alignment, which selects the highest-reward output among $N$ independent samples. Despite their widespread use, the fundamental limits of reward improvement under a KL budget remain poorly understood. We characterize the information-theoretic limits of KL-regularized
alignment by deriving the maximum achievable expected reward gain for a fixed KL-divergence budget. Our first result provides a closed-form expression for the optimal reward improvement, governed by a Jeffreys divergence term rather than the $\sqrt{\texttt{KL}}$ used in prior analyses. We further reformulate this expression as a covariance under the base model, yielding a practical
estimator that predicts achievable
alignment gains from base model samples alone. We extend our analysis to the proxy reward setting, showing that the gap between ideal and proxy
alignment (reward hacking) grows with the magnitude of reward error and when the KL penalty factor decreases. We then prove that reward ensembling mitigates reward hacking, providing a theoretical justification for this technique used in practice. Empirically, we compute the KL-reward Pareto frontier for two tasks for LMs, safety and summarization, and show that best-of-$N$ closely approaches the theoretical limit, while PPO and GRPO remain substantially suboptimal. Our theoretical results shed light on several empirically observed phenomena in the
alignment literature and suggest that algorithmic improvements are needed to achieve optimal
alignment without high inference costs.
Read the full abstract ▾
All authors Lucas Monteiro Paes, Natalie Mackraz, Barry-John Theobald, Federico Danieli