Chat
-
Detecting GPT-Assisted Writing Using Interpretable Stylometric Features
2026-09-22 16:44:26Rajesh Kumar, Nabeel Siddiqui, Alexander FuchsbergerAbstract
Distinguishing GPT-assisted from independently authored student writing has become a critical challenge in academia. This paper evaluates the discriminative capability of interpretable stylometric features extracted solely from submitted text. Using data from 90 participants who wrote both independently and with ChatGPT assistance, we evaluate eight machine learning classifiers while keeping data from the same participant together during validation. On the held-out test set, Random Forest achieved an ROC-AUC of 0.87 and an F1-score of 0.84, with False Positive and False Negative rates of 22.2% and 11.1%, respectively. SHAP analysis shows that lexical and grammatical characteristics drive the resulting predictions. The findings suggest that transparent, text-intrinsic features provide measurable signal for detecting GPT-assisted writing.
Abstract (translated)
URL
https://arxiv.org/abs/2609.26687
PDF
-
Safety Nudges: User-Facing Interventions for Real-Time AI Risk Awareness
2026-09-22 16:12:08Varshini Elangovan, James Wedgwood, Chhavi Yadav, William Agnew, Sauvik Das, Virginia SmithAbstract
Conversational AI systems can pose safety risks to their users such as hallucination, sycophancy, overconfidence, and anthropomorphism, but these risks are difficult for users to detect during everyday use. We introduce Safety Nudges, a browser-based tool that provides lightweight, in situ flags when concerning behavior is detected in chatbot conversations. We evaluated Safety Nudges in a two-week field study with 45 frequent chatbot users, collecting interaction logs, surveys, and feedback on individual nudges. Participants found the tool useful, clear, and minimally disruptive, with nearly all users reporting an increased awareness of potential AI harms, though we found that this improved awareness alone did not necessarily lead to discernible behavioral changes. Our results suggest that user facing safety nudges can complement model-level safeguards by helping people critically evaluate AI responses in context, while highlighting the importance of relevance, calibration, and user control in nudge design for conversational AI safety.
Abstract (translated)
URL
https://arxiv.org/abs/2609.26865
PDF
-
Capable yet Parsimonious: Extracting and Characterizing Hidden Chain-of-Thought in Frontier Models
2026-09-22 16:10:18Xiaoyu Luo, Tao Ren, Wenrui Yu, Xiao Li, Qiongxiu Li, Johannes BjervaAbstract
The rapid capability gains of frontier language models are widely attributed to improved reasoning abilities, yet this cannot be verified as raw CoT traces in closed-source systems are hidden. By registering a simple custom tool through a standard API feature, we induce frontier models to externalize intermediate reasoning. Because these traces may reflect post-hoc rationalization rather than genuine reasoning, we first evaluate against native CoT on open-source models and extend to closed-source frontier models including GPT-6 Astra. We find that the extracted reasoning matches native reasoning performance and substantially outperforms no-reasoning baselines, across competition mathematics, science, and code generation. We then characterize how frontier models structure their intermediate reasoning. Across token efficiency, reasoning-step types, and induced reasoning trees, we identify systematic differences in how models externalize, compress, and organize reasoning. We find that Astra exhibits token-efficient directed reasoning, selecting a correct trajectory earlier, while resolving elementary steps internally and externalizing only crucial reasoning. These findings provide a behavioral lens on frontier-model reasoning beyond benchmark scores.
Abstract (translated)
URL
https://arxiv.org/abs/2609.26637
PDF
-
FIRE: Failure-Informed Runtime Engineering for Reliable Language-Model Agents
Abstract
Language-model agents often reach a working solution and then fail to consistently deliver it. We study runtime policies: targeted natural-language instructions and action denials applied by the agent harness at states that preceded observed failures, without changing model weights or the user prompt. With this, keeping capability constant, we observe a meaningful unlock in delivered reliability. Across the complete 87-task Terminal-Bench 2.1 suite, with two attempts per task, policies increase repeated success (pass^2) in all three GPT-5.6 tiers: 50.6% to 54.0% for Luna, 55.2% to 60.9% for Terra, and 64.4% to 73.6% for Sol. Sol's best-of-two success changes by 1.2 points while repeated success rises by 9.2, showing that policies chiefly convert reachable solutions into dependable delivery. We further cover 14 tasks under Terra's frozen portfolio. Policy-guided Terra reaches 71.4%, compared with 64.3% for unassisted Sol, at about half the cost, demonstrating how engineering around models could unlock dependability for a use case. To isolate the mechanism we run a randomized five-arm experiment: real policies reach 61% on eligible tasks, versus 39% without a policy, 36% with a timing-matched sham, and 39 to 43% with generic verification or reconsideration. The intended corrective behavior appears in 22 of 24 coded policy attempts, against at most 14 in any other arm. Runtime policies are therefore a practical reliability layer: they make capabilities an agent already possesses substantially more repeatable.
Abstract (translated)
URL
https://arxiv.org/abs/2609.26048
PDF
-
Syndrome, Synergy, and Safety: Structured Reasoning and Knowledge-Driven Alignment for TCM Prescription Generation
2026-09-22 06:49:19Zheng Chen, ZhiCheng Du, Haoxuan Li, Peiwu QinAbstract
Applying large language models to Traditional Chinese Medicine (TCM) prescription generation reveals three clinically critical gaps: models produce end-to-end mappings without auditable reasoning following the li-fa-fang-yao paradigm (SR Gap), treat each encounter in isolation without follow-up adjustment via sui zheng jia jian (LA Gap), and fail to enforce absolute contraindication rules such as Shi Ba Fan (SC Gap). We propose a progressive four-stage framework (SFT $\to$ PG-CoT $\to$ Dynamic $\to$ K-RL) that addresses each gap: PG-CoT constrains CoT distillation under the li-fa-fang-yao paradigm to produce auditable diagnostic chains, Dynamic SFT models patient trajectories with explicit transition reasoning, and K-RL encodes deterministic pharmacological rules as rule-based DPO preference signals. Across 12 fine-tuned models and 6 zero-shot baselines, our framework substantially improves prescription quality over zero-shot baselines---with a 7B model (Mistral-7B) surpassing zero-shot GPT-5 on all three TCM evaluation metrics.
Abstract (translated)
URL
https://arxiv.org/abs/2609.25755
PDF
-
Interactive TTS: Dynamic Speaking Style Adaptation for Expressive Speech Synthesis
2026-09-22 05:25:02Wenjie Tian, Kangxiang Xia, Jingbin Hu, Xinfa Zhu, HangRui Hu, Ziyue Jiang, Kexin Huang, Ting He, Lei Xie, Jin XuAbstract
Dynamic speaking style adaptation in multi-turn multimodal interaction remains a major challenge for text-to-speech (TTS) systems. Existing context-aware TTS (CTTS) methods typically map dialogue context to speech in an end-to-end manner. Such implicit modeling makes contextual style decisions difficult to supervise, while the entanglement of style, timbre, and content often leads to weak instruction-following and severe timbre drift across turns. To overcome these limitations, we propose Interactive TTS, a dynamic, style-adaptive framework for contextually appropriate and speaker-consistent speech generation. Interactive TTS decouples the process by explicitly modeling contextual style decisions as executable instructions. To bridge the gap between style decisions and speech generation, we introduce Iterative Rejection Sampling Fine-Tuning (Iterative RSFT) and Context-Aware Direct Preference Optimization (CADPO), which significantly enhance instruction-following and align the generated speech with conversational contexts. Extensive experiments demonstrate that Interactive TTS outperforms state-of-the-art models on VStyle and SpeechParaling-Bench. Demo is available at this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2609.25707
PDF
-
Universal Fractal Natural Language Decision Map: Real-Time Edge Triage Across Heterogeneous Domains
2026-09-21 23:57:31Volkan Da\u{g}l{\i}, Zerrin Da\u{g}l{\i}, Da\u{g}han Da\u{g}l{\i}Abstract
Deploying Large Language Models for runtime operational triage incurs prohibitive latency (>100-500 ms), high VRAM requirements (>4-8 GB), and excessive energy dissipation. Extending Mandelbrot Fractal Neural Synthesis (Dagli et al., 2026), this paper presents the Universal Fractal Natural Language Decision Map, realized via the werr machine-native edge reflex runtime and the production answerr platform (this https URL). Operating entirely without stored weight tensors (0 Bytes VRAM), the engine synthesizes deterministic decisions---noul (Boolean), choice (categorical), and score (ordinal)---by dynamically modulating 24-byte coordinate seeds along the chaotic boundary of the Mandelbrot set and evaluating 4-quadrant escape dynamics. Drawing inspiration from biological System-One reflex arcs, the engine introduces: (i) an Auto-Seed Router with domain projector Phi_D yielding a +28.8% accuracy gain over linear baselines; (ii) an Information-Theoretic Acoustic Damping Filter grounded in token entropy and phonetic spectral density that insulates against prompt injections (0.0% empirical bypass; 95% Wilson CI: [0.0%, 30.8%]) while pruning escape iterations by 45.8% (accelerating throughput 2.5x to 3.31 ms latency); and (iii) an Organic Dynamic Calibration framework using O(1) Exponential Moving Average (EMA, alpha=0.03) and quadrant phase rotation to eliminate positional bias. Benchmarked on bare-metal infrastructure (this http URL) across 1,150+ verified decisions (3,200+ questions) and ranked World #1 on the independent JevBench suite (81.65%), the framework achieves 92.6% macro-accuracy (95% CI: [90.8%, 94.1%]) with 7.08 ms median CPU latency. We provide an OpenAI-compatible API (/v1/chat/completions) and demonstrate feasibility on microcontrollers and 32-byte EVM smart contracts.
Abstract (translated)
URL
https://arxiv.org/abs/2609.25498
PDF
-
Can We Still Trust Disaster Social Sensing? Empirical Evidence on Detecting AI-Generated Social Media Posts
2026-09-21 21:20:27Xiaoshan Zhou, Zaifu ZhanAbstract
Disaster social sensing converts public social-media posts into evidence for situational awareness and humanitarian needs, but generative artificial intelligence (AI) can produce plausible messages that resemble eyewitness reports. This study investigates whether text-based AI detectors can reliably distinguish human-authored from AI-generated disaster posts. We construct a dataset of 12,000 texts organised into 3,000 matched semantic units from nine disasters: original human posts (H0), minimally LLM-proofread human posts (H1), factual AI-generated posts based on the same verified facts (A0), and affectively framed versions of those AI posts (A1). A separate 6,000-text corpus from 42 events supports model selection and threshold calibration. We evaluate OSM-Det, Fast-DetectGPT, Binoculars, and direct large language model (LLM) judges across five model families, then test disaster-domain calibration, a frozen-encoder linear readout, paired transformation sensitivity, and dataset artifact controls. Across fourteen frozen cross-family configurations, AUROC is 0.402-0.517 and the best prospective recall at a calibration-derived low-false-positive operating point is 3.6%; OSM-Det reaches AUROC 0.521 and 10.4% recall at a realised 6.7% false-positive rate. A disaster-trained linear head reaches AUROC 0.817, but a seven-feature surface classifier reaches 0.784 on the H0-versus-A0 contrast, and neutralising identified surface asymmetries reduces the head from 0.733 to 0.594. The head also separates A0 from A1 even though provenance is unchanged. The results show that text-based detection is not reliable enough to serve as an operational trust gate; multimodal claims, accountable sources, and other contextual evidence should be rested on to safeguard trust in disaster social sensing.
Abstract (translated)
URL
https://arxiv.org/abs/2609.35821
PDF
-
TelecomGPT-R1: Unified Post-Training for Reasoning Across Heterogeneous Telecom Tasks
2026-09-21 19:48:02Bohao Wang, Chenwei Wu, Hang Zou, Yu Tian, Lina Bariah, Li Wei, Chongwen Huang, Yongliang Shen, Zhaoyang Zhang, Merouane DebbahAbstract
Large language models (LLMs) offer great potential to automate a broad range of telecom engineering tasks by reasoning over standards, network configurations, mathematical models, source code, and operational logs. However, existing telecom LLMs struggle to reliably reason across these diverse tasks and data types. General-purpose LLMs often lack reliable grounding in telecom-specific knowledge, while telecom-specialized models are typically developed for narrower task families and exhibit limited multi-task performance. To fill this gap, we introduce TelecomGPT-R1, a family of open source unified telecom reasoning models structured around four complementary axes: protocol, knowledge, modeling, and fault. We first develop an axis-aware data generation framework that refines coarse public telecom artifacts into verified question-answer pairs and high quality chain-of-thought (CoT) reasoning trajectories, yielding a training corpus containing 104,880 examples. Building on this corpus, supervised fine-tuning (SFT) instills telecom knowledge and evidence-grounded reasoning patterns to overcome the cold start barrier for reinforcement learning (RL). We then apply dynamic sampling policy optimization (DAPO) with task-routed rubric rewards to keep RL updates informative and stable across heterogeneous telecom reasoning tasks. These rewards decompose axis-specific CoT traces into verifiable reasoning units and combine grounded dense process credit with outcome correctness, allowing RL to learn generalizable problem solving behaviors from verifiable telecom evidence. We release the TelecomGPT-R1 models and a reproducible training recipe to support further community development. Evaluations on seven benchmarks of the GSMA Open Telco Leaderboard show that the open-source TelecomGPT-R1-27B achieves an 89.64% mean score, outperforming leading proprietary models, including GPT-5, Claude, and Gemini.
Abstract (translated)
URL
https://arxiv.org/abs/2609.25356
PDF
-
Learning to Plan in Human-Robot Collaboration: Multimodal Reinforcement Learning for Adaptive Interaction
2026-09-21 18:18:40Afagh Mehri Shervedani, Siyu Li, Natawut Monaikul, Bahareh Abbasi, Barbara Di Eugenio, Milo\v{s} \v{Z}efranAbstract
Robot assistants for older adults and people with disabilities need to perform collaborative tasks with users effectively. The core component of these systems is an interaction manager whose job is to observe and assess the task and infer the state of the human and their intent for the robot to choose the best course of action. Due to the sparseness of the data in this domain, the policy for such multimodal systems is often crafted by hand; as the complexity of interactions grows, this process is not scalable. This paper proposes a reinforcement learning (RL) approach to automatically generate the multimodal policy of the robot. Our system focuses on a realistic scenario where a robot assists a user in locating objects within a home environment, managing multimodal signals, including language and physical actions, to select the best action. In contrast to traditional dialog systems, our agent is trained with a simulator that uses human data and can deal with multiple modalities. We use a simple high-level reward function that needs no fine-tuning and enforce some preconditions to speed up the training process. A human study evaluating the system in a real-world setting demonstrates promising results, indicating high usability and effective task completion. This RL-based approach offers a scalable and interpretable alternative for designing interaction managers in multimodal human-robot collaborations.
Abstract (translated)
URL
https://arxiv.org/abs/2609.25274
PDF
-
Trains but Doesn't Learn: A Post-Training Delivery Benchmark for LLM Agents as Forward-Deployed Engineers
2026-09-21 18:01:24Weihang Ding, Junfei ZhanAbstract
Post-training is becoming a service (PTaaS): a customer hands an operator data and a goal, and a forward-deployed engineer (FDE) returns a fine-tuned, evaluated, and deployed model under a budget, a human-approval gate, and reproducibility requirements. Seating an LLM agent in the FDE seat raises a question existing benchmarks cannot answer: not whether an agent can raise a metric, but whether it can be trusted to deliver. We answer it on a governed delivery plane, where an agent drives ten stages and an oracle scores each stage from platform-recorded facts. The central silent failure is the run that trains but does not learn (TBDL): loss falls, every signal stays green, and the delivered model is no better than the base. An operator-run acceptance gate catches every such run before payment, and a detector calibrated on known-corrupted runs flags severe corruption mid-run. We ran four frontier agents (Claude Opus 5, GPT-5.6-luna, Gemini 3.7 Flash, DeepSeek V4-Pro) end to end on metered L40S, A100, and H200 GPUs across 8B to 70B open bases, certifying every scenario before scoring. We also ran a human FDE arm under the same oracle and compare every agent against it.
Abstract (translated)
URL
https://arxiv.org/abs/2609.25237
PDF
-
FinFIRST: Benchmarking Search Agents for Financial Information Retrieval, Sourcing and Traceability
2026-09-21 17:28:08Wenqing Wang, Haitao Xiang, Xinyi Zhao, Mingming Yin, Ying Zhong, Zhaoxin Huan, Qiheng Zhou, Jin Zhu, Xiaolu Zhang, Shi Chang, Jun ZhouAbstract
Financial search is a highly demanding task for LLM agents, requiring not only a correct final answer but also temporally valid information retrieval, authoritative source selection, entity and period alignment, unit and definition consistency, and verifiable evidence for all conclusions. Existing benchmarks predominantly evaluate only the final answer, making it difficult to localize errors or assess whether an answer is well-founded. To address this gap, we introduce FinFIRST (Financial Information Retrieval, Sourcing and Traceability), the first financial benchmark to jointly evaluate answers and supporting evidence through atomic rubrics. FinFIRST comprises 123 expert-authored tasks spanning a graduated difficulty spectrum, constructed from aggregate patterns of real-world financial scenarios through an 18-field taxonomy, a six-axis coverage blueprint, a registry of 138 financial sources, contributions from over 50 finance experts, and a six-stage quality-control pipeline. Each task is accompanied by an evidence-grounded reference package decomposed into atomic criteria across three dimensions: raw-information acquisition, source verification, and computation and answer formation. We evaluate 15 model configurations under a unified tool setting. Claude-Opus-5 achieves the highest atomic score of 87.59%, while GPT-5.6-Sol attains the highest strict pass rate of 71.54%. Computation and answer formation consistently lag behind raw-information acquisition across systems. FinFIRST retains final-answer correctness as the primary objective while making the supporting research process measurable, verifiable, and diagnosable.
Abstract (translated)
URL
https://arxiv.org/abs/2609.25192
PDF
-
Linguistic Features for Interpretable Textual Entailment
2026-09-21 17:26:30David Torres-Moreno, Jorge Hermosillo-Valadez, Asela Reig-AlamilloAbstract
Despite the success of neural models in natural language processing, their black-box nature limits interpretability and conceals the linguistic phenomena underlying their predictions. We present SLITE, an explainable hybrid model for Recognizing Textual Entailment that integrates two complementary layers of semantic analysis: a structural-relational layer, based on semantic compatibility and incompatibility between compositional entities, and a distributional-informational layer, based on structured patterns of information change between embedding-based representations of the premise and the hypothesis. We propose 17 features that combine entity-level semantic relations, polarity-sensitive lexical matching, and alignment measures over semantic sub-representations of the similarity matrix, including measures based on entropy and transfer entropy. A logistic regression trained on these features achieves an accuracy of 83% on three-class SICK and 96% on SICK-CE, outperforming IsoLex by 4 percentage points and falling within 2 percentage points of RoBERTa with a fraction of its computational complexity. Ablation studies and SHAP analysis confirm that structural-relational features are the primary drivers of classification, while distributional-informational features provide essential complementary contributions, particularly for detecting neutrality and contradiction. Our results demonstrate that further exploration of hybrid approaches is a viable and scientifically productive alternative to massive neural architectures, and we hope they will strengthen the dialogue between linguistic theory and computational modeling of inference
Abstract (translated)
URL
https://arxiv.org/abs/2609.24932
PDF
-
SLICEChat: Progressive In-Encoder Token Pruning for Whole-Slide Pathology Language Models
2026-09-21 16:59:06Ali Kerem Bozkurt, Baris Cem Bakay, Ibrahim Kulac, Cigdem Gunduz-Demir, Erkut Erdem, Aykut ErdemAbstract
Whole-slide pathology images (WSIs) contain gigapixel-scale visual content, creating a major scalability challenge for slide-level multimodal large language models (MLLMs). Existing approaches process thousands of patch tokens and typically apply compression only after slide encoding, leaving multimodal attention computationally expensive. We introduce SLICEChat, a slide-level MLLM that integrates progressive token pruning within a hybrid Mamba--Transformer slide encoder. Mamba layers enable efficient long-range propagation, while Transformer layers preserve global interactions as the sequence is progressively shortened. Between stages, language-supervised, region-aware pruning removes spatially coherent low-utility regions under a controlled keep-rate schedule, producing compact slide representations before multimodal fusion. On SlideBench VQA, SLICEChat achieves 79.84% accuracy on TCGA and 59.09% on BCNB cohorts, outperforming prior slide-level pathology MLLMs, and achieves the highest overall WSI-Bench metrics. It also provides competitive memory usage and the inference latency among the evaluated models. These results demonstrate accurate and computationally efficient multimodal reasoning over gigapixel WSIs.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24894
PDF
-
Small-world Networks of Agents Brainstorm AI Risks to Support Ideation
2026-09-21 16:32:57Ke Zhou, Edyta Bogucka, Daniele QuerciaAbstract
The ideation phase of participatory AI risk assessment often starts with a blank slate or a limited list of predefined risks, making it difficult to surface indirect or systemic harms. To address this limitation, we propose a three-stage ideation support tool. The tool complements participatory AI, rather than replacing it, and helps focus later engagement with affected communities. First, it dynamically discovers stakeholders depending on the given AI use and recursively expanding outward, allowing overlooked or indirect stakeholders to emerge. Second, it simulates these stakeholders with LLMs, connecting them into a network of a given topology, and having them ideate about risks. Third, it prioritizes risks using network centrality measures. In an initial evaluation, we found that betweenness centrality run through agents connected in a small-world network works best as it elevates risks raised by stakeholders who bridge disconnected groups, surfacing novel, systemic harms that traditional methods often miss. On an AI chatbot companion use case, this approach increased the novelty of the identified risks by approximately 1.1 points over single LLM brainstorming, and by 0.5 points over agentic LLM brainstorming, measured on a normalized five-point Likert scale, without reducing the plausibility or severity of the identified risks. To test whether our framework helps a human-led ideation session using the Futures Wheel approach, we divided 11 teams of non-western young chatbot users into two types: control (team) and treatment (team) in a participatory AI risk assessment. The control teams started from a list of risks generated by the 45 AI practitioners in the initial evaluation; the treatment teams started from a list generated by our framework. The treatment teams identified more risks overall, and more systemic, human-computer interaction, and environmental risks.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24859
PDF
-
SPECTRA: Adaptive Execution of Speculative Decoding on a Runtime-Reconfigurable Tiled Architecture
2026-09-21 16:25:24Gabriele Tombesi, William Baisi, Je Yang, Elisavet Lydia Alvanaki, Kevin Lee, Michael Lippe, Biruk Seyoum, Luca P. CarloniAbstract
LLM inference on edge devices is constrained by computational and memory resources, making efficient autoregressive decoding challenging. Speculative decoding alleviates this bottleneck by generating tokens with a smaller draft model and verifying multiple tokens in parallel with a batched target model pass. However, verification introduces a runtime-dependent intermediate regime between memory-bound general matrix-vector (GEMV) operations in decoding and compute-bound general matrix-matrix (GEMM) operations in prefill, as its arithmetic intensity varies with speculation length and acceptance rate. We present SPECTRA, a runtime-reconfigurable tiled architecture that sustains high utilization across the full speculative decoding pipeline. Within each tile, the compute engine switches between systolic execution for GEMMs and vector-lane execution for GEMVs. Across tiles, SPECTRA dynamically adapts computation parallelism by selecting tile count, kernel partitioning, and communication pattern. Both tile-level and system-level reconfiguration operate on a per-kernel basis, enabling efficient execution across these diverse regimes. Evaluated on a 20-tile FPGA prototype across the Pythia, SmolLM2, and GPT-2 families, SPECTRA achieves up to $2.09\times$ speedup from tile-level reconfiguration and a further $1.25\times$ gain from system-level adaptability over fixed designs.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24847
PDF
-
PrismGPT: Proxy-Guided Learning for Region-Aware Photo Editing with Self-Synthesized Reasoning
2026-09-21 15:35:53Ke Zhao, Hue Nguyen, Abhijith Punnappurath, Zhongling Wang, Iqbal Mohomed, Michael S. BrownAbstract
Professional photo finishing relies on both global adjustments and region-specific local edits guided by semantic masks, yet current automated methods handle this workflow only partially. We present PrismGPT, a Vision-Language Model (VLM) framework that produces structured, region-aware editing plans from a single input image without relying on commercial black-box tools. Training a VLM to simultaneously diagnose aesthetic deficiencies at both global and local levels while predicting precise editing parameters is challenging due to the vast combinatorial decision space. We address this through proxy-guided learning: two simpler proxy tasks -- operation decomposition and region-aware aesthetic ranking -- teach the foundational skills the model needs, while a competence-based dynamic scheduler automatically rebalances the multi-task training ratio, progressively shifting emphasis from the proxy tasks to the primary editing task as each skill is mastered. Crucially, all reasoning traces used for supervised fine-tuning are self-synthesized by the same base model, eliminating the need for a stronger external teacher. Experiments on MIT-Adobe FiveK and SPIRE, a new professionally retouched benchmark we introduce, show that PrismGPT achieves state-of-the-art results while using only ~6% of the training data compared to the previous best method.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24768
PDF
-
Construting Reverse Thinking: Developing Large Language Models' Reverse Thingking Ability
2026-09-21 15:29:46Xin Liu, Yunhai Li, Chunfu Jia, Ziliang Chen, Jisen SongAbstract
When facing complex problems, humans tend to try various ideas for different issues. Human thinking patterns exhibit remarkable flexibility in adapting to diverse scenarios. GPT-o1, GPT-o3, and DeepSeek-R1 adopt long chain-of-thought models to address complex problems by increasing reasoning depth, which default to a forward reasoning mode. We conducted statistical analysis on the accuracy of different mathematical problem datasets on models of different scales, and found five reasons for errors: Insufficient solution-space coverage, Computational mistakes, Unverified assumptions, Ignoring constraint conditions, Maximum response length limitation. To address the above issues, we proposed a backward reasoning pattern construction method aimed at enhancing the model's reverse thinking ability and dynamic adaptability. First, we constructed an easy-hard two-stage Math dataset for training large models and gradually improving their inference ability at different difficulty levels. The dataset contains forward reasoning paths as well as backward reasoning paths. And a two-stage supervised fine-tuning process is applied to progressively train the model's backward reasoning capability. Furthermore, a fine-grained reward mechanism is developed, employing smoothed reward signals to strengthen the model's ability to autonomously select thinking modes during the reasoning process, thereby avoiding reward hacking. A linear-decay balanced sampling strategy is designed to maintain a balance between forward and backward reasoning path samples during training, enabling the model to converge quickly and stably. Experimental results show that our method significantly improves reasoning efficiency and accuracy in tasks such as mathematical proofs, offering a flexible and efficient reasoning paradigm for solving complex problems.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24760
PDF
-
LLM-based Conversational AI Knowledge Assistant for MyBuddy Humanoid Robot
2026-09-21 15:11:06Hanxiao ChenAbstract
Humanoid robots are increasingly being popular and developed for human-centered applications, yet their ability to provide intelligent conversations and natural interactive knowledge assistance remains constrained by traditional rule-based dialogue systems, pre-defined responses and limited knowledge repositories. Large language models (LLMs) have emerged as a powerful foundation for enabling natural, adaptive, and context-aware Human-Robot Interaction (HRI), which provides a significant opportunity to address such limitations by enabling robots to understand natural speech language, reason over complicated queries, maintain high-quality conversational context, and generate knowledge-rich responses. In this work, we originally present and implement an LLM-based versatile Conversational AI Knowledge Assistant for the Raspberry-Pi-powered 13-Axis MyBuddy humanoid robot, which integrates LLM-driven language understanding and AI reasoning with real-time speech recognition, knowledge retrieval via extensible access of internet engines (e.g., Wikipedia, arXiv), flexible dialogue management, and natural speech synthesis to enable much more intelligent multi-turn continuous conversations and advanced emotional-support Human-Robot Interaction.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24742
PDF
-
Reasoning Topology Matters: A Controlled Study of LLM-Based Cybersecurity Analysis
2026-09-21 14:52:30Jiling Zhou, Aisvarya Adeseye, Antti Hakkala, Seppo Virtanen, Jouni IsoahoAbstract
Large Language Models (LLMs) are increasingly used in cybersecurity, where accurate analysis often requires multi-step and context-dependent reasoning over complex and heterogeneous data. However, existing prompting approaches typically focus on eliciting reasoning without explicitly considering how intermediate reasoning steps are structurally organized. We introduce Security Reasoning Topology, which models reasoning through three representative structures: Linear, Branching, and Graph. To evaluate their effects, we conduct controlled experiments on three cybersecurity datasets covering MITRE ATT&CK network traffic, cyber threat intelligence (CTI), and CVE vulnerability analysis. We evaluate multiple LLMs, including Llama 2 (7B, 13B, 70B), GPT-5.1, and Mistral Large 3, while keeping task inputs consistent and controlling reasoning structure through system-level prompting. Results show that reasoning topology substantially affects performance: Graph reasoning achieves the highest overall accuracy, improving over few-shot prompting by 9.8-12.2 percentage points across datasets, while Branching provides a strong intermediate solution. The results further show that the effect of reasoning topology remains consistent across model families and scales, highlighting reasoning topology as an important design factor for LLM-based cybersecurity analysis.
Abstract (translated)
URL
https://arxiv.org/abs/2609.24710
PDF
