arXiv subjects

Siyu Song

Publications and source records attributed to Siyu Song.

16 recordsLinked to original sources

arXiv2026

Text as an Index for Zero-Shot Cross-Domain Knowledge Tracing

Knowledge tracing (KT) predicts a student's next response from their history. New platforms often have question texts but no response logs for training, motivating zero-shot cross-domain KT. Existing transfer and cold-start methods use question semantics, but cross-platform semantic discrepancies limit transferability. We propose TIZSKT, which separates behavioral sequence modeling from domain-local historical evidence construction and uses text as an index rather than a directly transferable predictive representation. The Mastery-Dynamics Encoder learns correctness dynamics without question identifiers or text embeddings. Domain-Local Question Addressing builds fixed question addresses using a frozen text encoder and an unsupervised per-domain projection. The Text-Addressing Kernel uses these addresses to aggregate the current student's previous outcomes into a question-specific historical correctness summary. Cross-Segment Continuation retains access to earlier responses without extending the behavioral encoder's input sequence. Evidence-Volume Read-Out preserves the total kernel mass discarded by normalization and combines its log-compressed value with the behavioral state and historical correctness summary to predict the next response. All trainable parameters are learned from source domains; target inference uses public question texts and the current student's observed history without target-domain training. Leave-one-domain-out experiments on three public datasets show that TIZSKT achieves the highest mean AUC and accuracy among evaluated baselines on all target domains. Ablations confirm the contributions of text addressing and cross-segment continuation.

cs.CL↗

arXiv2026

The Predictive--Optimization Gap in GuanDan Self-Play

Outcome-predictive signals are often treated as promising optimization targets in reinforcement learning, but predictive validity and optimization utility are different claims. We expose this distinction in GuanDan, a multi-round imperfect-information game in which the implication of a round outcome depends on the public match state. Held-out analysis shows that state--gain interactions provide incremental predictive value beyond a constant signed-gain effect, establishing that match progress is predictively meaningful. We then use changes in a frozen match-win predictor, $ΔΦ$, as a controlled round-level supervision intervention in a matched DanZero-style Monte Carlo self-play protocol. The intervention can succeed in an individual lineage: a selected progress-augmented checkpoint substantially outperforms its matched Baseline on a locked evaluation. Yet the method-level result reverses this impression: none of five preregistered replication seeds is positive, the pooled ten-seed interval includes zero, and across-iteration effects repeatedly change sign without a common late-stage advantage. These results identify a predictive--optimization gap: under the studied learner, a signal can describe outcome-relevant progress while failing to provide a reliably beneficial optimization direction.

cs.LG↗

arXiv2026

EduAlign: Aligning Educational LLMs for Helpfulness, Personalization, and Creativity

Educational language models should pursue three complementary objectives: helpfulness through responsible guidance, personalization to learner needs, and creativity that supports exploration. General-purpose models do not consistently satisfy these educational requirements, motivating explicit alignment with all three objectives. We present EduAlign, a framework that combines educational reward modeling with multi-objective reinforcement learning. To obtain training signals for these goals across educational scenarios, we develop a rubric-guided data synthesis pipeline that iteratively refines scoring criteria and generation prompts for each subject, grade band, and task type. The resulting supervision trains a multidimensional reward model, Edu-RM, to predict helpfulness, personalization, and creativity scores. To jointly optimize these educational objectives, we introduce Ideal-point Policy Optimization (IdealPO), a multi-objective reinforcement learning algorithm. IdealPO combines ideal-point marginal credit with continuous conflict weighting to guide policy updates that account for uneven progress and conflicting contributions across objectives. Experiments with 7B and 32B models show that EduAlign achieves the highest mean ratings on all three dimensions from both education experts and a model judge, alongside the highest aggregate scores on EduBench, ELMES, and EduValues among the compared methods. Reward-model validation and component ablations further support the effectiveness of the framework.

cs.LG↗

arXiv2026

CORE: Conflict-Oriented Reasoning Elimination for Verifiable Language-Model Search

Test-time reasoning systems often respond to failure by restarting or revising the latest step, even when an earlier decision caused the error. We introduce CORE, a search controller that requests a certified conflict core from a verifier, backjumps to the latest decision in that core, and caches the conflict to avoid repeating it. Under sound verification, finite branching and depth, and exhaustive proposals, the uncapped search is complete and never prunes a valid solution. On 2,000 planted graph-coloring instances with matched proposals and an exact verifier, CORE reduces median verifier calls by 39.8% at 30 variables and 35.0% at 36 variables relative to chronological repair; caching further improves on backjumping alone. Across five reasoning tasks, CORE achieves 75.9% mean success with Qwen2.5-7B-Instruct and 84.2% with Qwen3-8B, compared with 72.5% and 81.8% for Tree of Thoughts. It also uses fewer verifier calls and generated tokens on both backbones. These results show the value of using certified failure explanations to direct language-model search.

cs.CL↗

arXiv2026

AnthroDial: Benchmarking LLM Anthropomorphism in Autonomous Social Interaction

Large language models (LLMs) are increasingly deployed as social agents, yet credible human-like interaction requires more than fluent responses or persona consistency. Agents must autonomously decide whether, when, and how to communicate while adapting to evolving contexts, goals, and relationships. Existing research, however, lacks a unified approach to enabling, evaluating, and improving such capabilities in continuous, open-ended interaction. We introduce AnthroDial, a unified framework for developing anthropomorphic social agents from three complementary aspects: MindFlow, a lightweight interaction harness that enables autonomous, asynchronous, and adaptive communication through a dynamic Mind Buffer; CAPS-Eval, a theory-grounded framework for evaluating cognitive, affective, and behavioral dimensions of anthropomorphic interaction; and a scalable training paradigm that combines SEEDS for environment expansion with DiAPO for adaptive capability optimization. We further construct evaluation datasets covering everyday communication, game interaction, and long-horizon character interaction. Extensive experiments across diverse models and scenarios demonstrate improved interaction autonomy and naturalness, validate the reliability, discriminativeness, and agreement with human rankings of CAPS-Eval, and confirm the effectiveness of our training paradigm. Together, these components provide a unified framework for developing credible human-like social agents in open-ended interaction.

cs.CL↗

arXiv2026

Deadline-Aware Adaptive Prefill Chunking for Efficient Large Language Model Serving

Continuous batching improves large language model (LLM) serving throughput, but long prompt prefills can delay decode iterations and violate inter-token latency objectives. Chunked prefill mitigates this interference, yet its chunk size is normally fixed: small chunks protect decode latency but repeatedly pay launch overhead, while large chunks improve prefill efficiency but create latency spikes. We introduce SLOWeave, an online scheduling method that selects the largest prefill chunk predicted to finish before the earliest active decode deadline. The decision requires no workload-specific chunk-size tuning and is computed by a logarithmic-time search over a monotone iteration-cost model. We prove that, whenever a decode-only iteration is feasible and the cost predictor is accurate, SLOWeave maximizes immediate prefill progress among decisions that preserve every active request's next-token deadline. We evaluate the method in a reproducible event-driven simulator and an iteration-level GPU runtime across chat, mixed-context, long-context, and bursty workloads. Under a 25ms time-per-output-token objective, SLOWeave improves goodput over the strongest fixed-chunk baseline by 39% on mixed requests and 38% on long-context requests. Under a stricter 10ms objective, the gains rise to 3.3$\times$ and 2.4$\times$, respectively. These results isolate adaptive chunk sizing as a useful serving primitive and provide an implementation-ready controller for integration with iteration-level LLM runtimes.

cs.CL↗

arXiv2026

Beyond Direct Answering: Aligning Educational LLMs as Socratic Guides via Heuristic Reinforcement Learning

Large language models (LLMs) deployed in educational settings often behave as direct answerers: they disclose target concepts in the opening turn instead of guiding students through progressive inquiry, as Socratic pedagogy prescribes. We present HeuristicEdu, a two-phase pipeline that aligns Qwen2.5-7B toward Socratic tutoring via supervised warm-up and Group Relative Policy Optimization (GRPO). Training uses SocraticEdu, 797 multi-turn Chinese children's science dialogues reconstructed from a live platform, with a heuristic reward over cognitive depth (R_cog), curiosity engagement (R_eng), and directness (R_dir), together with a K_query correction for student-introduced terms. We introduce Scaffolding Effectiveness (SE) and Conversation Depth (CD) to evaluate outcomes beyond surface fluency. On 30 held-out questions, the best GRPO variant improves SE from 30.0% to 63.3% and lowers keyword leakage from 30.0% to 13.3%. Notably, this best variant omits the directness penalty during optimization, suggesting that explicit anti-leakage terms can conflict with gradient-based behavioral alignment. An unaligned Qwen-72B baseline reaches 0% SE and 96.7% leakage, showing that scale alone does not induce Socratic behavior.

cs.CL↗

arXiv2026

Toward Anthropomorphic Dialogue: A Closed-Loop Framework for Human-Like Chat Generation, Evaluation, and Preference Alignment

Human-like private chat requires more than fluent response generation: a system must preserve persona, relationship, memory, bounded knowledge, medium-specific timing, and a coherent multi-turn arc. We present AnthroDial, a closed-loop framework that formulates anthropomorphic dialogue as a joint problem of system architecture, executable evaluation, and diagnostic alignment. It combines (1) a role-conditioned scheduled dialogue runtime with persona and scenario cards, long-term memory, virtual time, and single-draft message decisions; (2) an executable benchmark with an L0 validity gate, five per-turn dimensions, and five dialogue-level dimensions; and (3) a post-training pipeline that filters 16,436 scheduled-decision examples for SFT and applies GRPO with a cognitive-diagnostic, ZPD-aware reward. The reward maintains Kalman-filtered capability estimates for each behavioral dimension, upweights dimensions with larger capability deficits, and uses rollout scores as task-level ZPD matches to focus optimization on learnable weak skills. On a benchmark with 55 personas, 50 scenarios, 50 persona-scenario bindings, and 100 role-conditioned cases per model, we evaluate 16 systems spanning frontier baselines, open models, thinking/no-think variants, and SFT/RL ablations. The strongest non-trained baseline reaches 32.00% strict ACC, while Qwen3.6-27B-SFT+RL reaches 39.00% strict ACC and a 98.5 overall score. In the 9B no-think setting, SFT and RL improve strict ACC from 0.00% to 13.00% and 18.37%. These results show that anthropomorphic dialogue benefits when generation, evaluation, and reward shaping share the same behavioral dimensions.

cs.AI↗

arXiv2026

Elmes*: Automated Construction of Fine-Grained Evaluation Rubrics for Large Language Models in Long-Tail Educational Scenarios

Evaluating large language models (LLMs) for education requires measuring how models teach, not only what they know. Existing benchmarks emphasize domain-general correctness or depend on manually designed rubrics that scale poorly to long-tail pedagogical scenarios. We introduce Elmes*, an end-to-end framework for constructing, refining, and applying fine-grained scenario-specific rubrics. Elmes* combines a declarative multi-agent engine for teacher--student--judge interactions with SceneGen, a self-evolving module that co-optimizes evaluation criteria and test data from expert-defined pedagogical dimensions. Using Elmes*, we build Edu-330, covering 330 scenarios across 11 subjects, 3 grade bands, and 10 task types, with over 1{,}000 second-level indicators. Experiments on Edu-330 and four expert-authored gold-standard scenarios show that educational capability is multidimensional: top-tier LLMs differ mainly in creativity and values integration, knowledge-strong models may fail at Socratic scaffolding, and the education-specialized InnoSpark achieves the best human-evaluated average score. LLM judges preserve human-comparable rankings with much lower scoring variance, but exhibit judge-specific biases such as self-preference. Ablations show that expert-scored few-shot anchoring improves human--LLM alignment, while reasoning enforcement and greedy decoding are model-dependent. Elmes* thus provides scalable diagnostic infrastructure for pedagogically grounded LLM evaluation.

cs.LG↗

arXiv2026

Automating Skill Acquisition through Large-Scale Mining of Open-Source Agentic Repositories: A Framework for Multi-Agent Procedural Knowledge Extraction

The transition from monolithic large language models (LLMs) to modular, skill-equipped agents represents a fundamental architectural shift in artificial intelligence deployment. While general-purpose models demonstrate remarkable breadth in declarative knowledge, their utility in autonomous workflows is frequently constrained by insufficient specialized procedural expertise. This report investigates a systematic framework for automated acquisition of high-quality agent skills through mining of open-source repositories on platforms such as GitHub. We focus on the extraction of visualization and educational capabilities from state-of-the-art systems including TheoremExplainAgent and Code2Video, both utilizing the Manim mathematical animation engine. The framework encompasses repository structural analysis, semantic skill identification through dense retrieval, and translation to the standardized SKILL.md format. We demonstrate that systematic extraction from agentic repositories, combined with rigorous security governance and multi-dimensional evaluation metrics, enables scalable acquisition of procedural knowledge that augments LLM capabilities without requiring model retraining. Our analysis reveals that agent-generated educational content can achieve 40\% gains in knowledge transfer efficiency while maintaining pedagogical quality comparable to human-crafted tutorials.

cs.AI↗

arXiv2026

Scaling Laws for Educational AI Agents

While scaling laws for Large Language Models (LLMs) have been extensively studied along dimensions of model parameters, training data, and compute, the scaling behavior of LLM-based educational agents remains unexplored. We propose that educational agent capability scales not merely with the underlying model size, but through structured dimensions that we collectively term the Agent Scaling Law: role definition clarity, skill depth, tool completeness, runtime capability, and educator expertise injection. Central to this framework is AgentProfile, a structured JSON-based specification that serves as the mechanism enabling systematic capability growth of educational agents. We present EduClaw, a profile-driven multi-agent platform that operationalizes this scaling law, demonstrating its effectiveness through the construction and deployment of 330+ educational agent profiles encompassing 1,100+ skill modules across K-12 subjects. Our empirical observations suggest that educational agent performance scales predictably with profile structural richness. We identify two complementary scaling axes -- Tool Scaling and Skill Scaling -- as future directions, arguing that the path to more capable educational AI lies not solely in larger models, but in stronger structured capability systems.

cs.AI↗

arXiv2025

AutoSynth: Automated Workflow Optimization for High-Quality Synthetic Dataset Generation via Monte Carlo Tree Search

Supervised fine-tuning (SFT) of large language models (LLMs) for specialized tasks requires high-quality datasets, but manual curation is prohibitively expensive. Synthetic data generation offers scalability, but its effectiveness relies on complex, multi-stage workflows, integrating prompt engineering and model orchestration. Existing automated workflow methods face a cold start problem: they require labeled datasets for reward modeling, which is especially problematic for subjective, open-ended tasks with no objective ground truth. We introduce AutoSynth, a framework that automates workflow discovery and optimization without reference datasets by reframing the problem as a Monte Carlo Tree Search guided by a novel dataset-free hybrid reward. This reward enables meta-learning through two LLM-as-judge components: one evaluates sample quality using dynamically generated task-specific metrics, and another assesses workflow code and prompt quality. Experiments on subjective educational tasks show that while expert-designed workflows achieve higher human preference rates (96-99% win rates vs. AutoSynth's 40-51%), models trained on AutoSynth-generated data dramatically outperform baselines (40-51% vs. 2-5%) and match or surpass expert workflows on certain metrics, suggesting discovery of quality dimensions beyond human intuition. These results are achieved while reducing human effort from 5-7 hours to just 30 minutes (>90% reduction). AutoSynth tackles the cold start issue in data-centric AI, offering a scalable, cost-effective method for subjective LLM tasks. Code: https://github.com/bisz9918-maker/AutoSynth.

cs.LG↗

arXiv2025

EA4LLM: A Gradient-Free Approach to Large Language Model Optimization via Evolutionary Algorithms

In recent years, large language models (LLMs) have made remarkable progress, with model optimization primarily relying on gradient-based optimizers such as Adam. However, these gradient-based methods impose stringent hardware requirements, demanding high-concurrency, high-memory GPUs. Moreover, they require all neural network operations to be differentiable, thereby excluding many promising non-differentiable architectures from practical use. To address these limitations, we propose EA4LLM, an evolutionary algorithm for optimizing LLMs, and, for the first time, empirically verify full-parameter optimization from the pretraining stage across model sizes ranging from 0.5B to 32B. We conduct extensive experiments and provide key insights into how evolutionary algorithms can effectively optimize neural networks. Our work challenges the prevailing assumption that gradient-based optimization is the only viable approach for training neural networks. It also holds significant potential to reduce the computational cost of training large language models, thereby enabling groups with limited computational resources to participate in deep learning research.

cs.AI↗

arXiv2025

First-Principles Investigation of Auxetic Piezoelectric Effect in Nitride Perovskites

The recently reported auxetic piezoelectric effect, which acts as the electrical counterpart of the negative Poisson's ratio, is of significant technical importance for applications in acoustic wave devices. However, this electric auxetic effect has not yet been reported in perovskite systems. In this work, we employ first-principles calculations to investigate the piezoelectric properties of six polar nitride perovskites with the chemical formula ABN3 (A = La, Sc, Y; B = W, Mo). Among these, all compounds except ScMoN3 exhibit the auxetic piezoelectric effect, which is characterized by an unusually positive transverse piezoelectric coefficient, along with a positive longitudinal piezoelectric coefficient. This behavior is in sharp contrast to previously reported results in HfO2, where both the longitudinal and transverse piezoelectric coefficients are negative. These unusual positive transverse piezoelectric coefficients originate from the domination of the positive internal-strain contribution. We further confirm the auxetic piezoelectric effect with finite electric field calculations. Our research enriches the understanding of the piezoelectric properties of nitride perovskites and provides a new compositional space for the design of novel auxetic piezoelectric materials.

cond-mat.mtrl-sci↗

arXiv2025

Enhanced Charge Transport in A-site Ordered Perovskite Derivatives A2A'Bi2I9 (A = Cs; A'= Ag, Cu): A First-Principles Study

Recent experiments have synthesized Cs2AgBi2I9 by partially substituting Cs+ with Ag+ at the A-site of Cs3Bi2I9, resulting in enhanced charge transport properties compared to Cs3Bi2I9. However, the atomic-scale mechanisms behind this enhancement remain unclear. In this work, we investigate the carrier transport mechanisms in Cs2A'Bi2I9 (A' = Ag, Cu) using first-principles calculations and Boltzmann transport calculations. Our results reveal that A-site ordered Cs2A'Bi2I9 exhibits carrier mobilities that are 3-4 times higher than those of Cs3Bi2I9 within the 100-500 K temperature range. We identify polar phonon scattering as the dominant mechanism limiting mobility. Furthermore, the enhanced out-of-plane carrier mobility in Cs2A'Bi2I9, particularly between 100 and 200K, leads to reduced mobility anisotropy. These improvements are mainly due to the shorter A'-I bond lengths and increased Ag+/Cu+ s-I p orbital coupling. Notably, substitution with Cu+ results in a further reduction in the band gap and enhanced hole mobility compared to Ag+ substitution in Cs3Bi2I9. Further analysis reveals that the significant increase in carrier mobility in Cs2A'Bi2I9 can be largely explained by the smaller carrier effective masses (m*) and weaker Fröhlich coupling strengths (α), resulting in a lower polar mass α(m*/me), compared to Cs3Bi2I9. Our study provides valuable insights into the transport properties of Bi-based perovskite derivatives, paving the way for their future applications in optoelectronic devices.

cond-mat.mtrl-sci↗

arXiv2024

GrootVL: Tree Topology is All You Need in State Space Model

The state space models, employing recursively propagated features, demonstrate strong representation capabilities comparable to Transformer models and superior efficiency. However, constrained by the inherent geometric constraints of sequences, it still falls short in modeling long-range dependencies. To address this issue, we propose the GrootVL network, which first dynamically generates a tree topology based on spatial relationships and input features. Then, feature propagation is performed based on this graph, thereby breaking the original sequence constraints to achieve stronger representation capabilities. Additionally, we introduce a linear complexity dynamic programming algorithm to enhance long-range interactions without increasing computational cost. GrootVL is a versatile multimodal framework that can be applied to both visual and textual tasks. Extensive experiments demonstrate that our method significantly outperforms existing structured state space models on image classification, object detection and segmentation. Besides, by fine-tuning large language models, our approach achieves consistent improvements in multiple textual tasks at minor training cost.

cs.LG↗