MARS is a scalable knowledge‑graph question‑answering system that integrates large language models with knowledge graphs without requiring model fine‑tuning. It uses a structured retrieval process that links question entities to the graph, iteratively fetching next‑hop information and deciding when to stop and generate a SPARQL query. Evaluated on three KGQA benchmarks—including multilingual settings—MARS matches state‑of‑the‑art performance while remaining efficient and scalable.
By Nikit Srivastava, Daniel Vollmers, Ren\'e Speck, Nikolaos Karalis, Hamada M. Zahera, Axel-Cyrille Ngonga Ngomo
The paper introduces CADRE, a family of causal diffusion representation autoencoders designed for video world models that balance spatio‑temporal compression. By training the same world model across different CADRE variants in manipulation and navigation tasks, the authors map the latency‑quality trade‑off and uncover that strong compression favors diffusion decoders, temporal compression is more effective in the autoencoder for manipulation, and spatial compression is preferable when the camera moves. The study also reveals that coarse spatial tokens degrade motion perception more than image quality when the camera is static.
By Adrien Ramanana Rahary, Yutong Bai, Nicolas Dufour, Patrick P\'erez, David Picard
AutoExpert proposes a new benchmark for automating 3D LiDAR annotation using expert‑crafted guidelines, redefining evaluation protocols and re‑annotating nuScenes and PandaSet datasets. The pipeline combines 2D detection and segmentation in RGB images, lifts these detections into 3D with known sensor poses, and generates 3D cuboids, improving 3D detection mAP from 12.1 to 25.4 on the AutoExpert‑nuScenes benchmark.
By Yechi Ma, Wei Hua, Shu Kong
The paper introduces Effect-Marginal Policy Optimization (EMPO), a method that addresses credit-update misalignment in multi-turn reinforcement learning for large language model agents. EMPO aligns policy updates with execution-level equivalence by using an effect-level surrogate that couples updates across different textual responses that produce the same effect. Experiments on ALFWorld, WebShop, and Sokoban demonstrate that EMPO improves both task performance and learning efficiency compared to strong baselines.
By Shaojin Ma, Ruitao Wang, Pengyi Li, Jie Yang, Liangyu Chen, Gangyi Zhao, Hongyao Tang, Yan Zheng, Jianye Hao
The Data Differentiator Framework (DDF) is a composite scoring method that evaluates candidate data assets by combining a mutual‑information redundancy penalty, a cost‑normalized marginal utility score from an ablation study, and explicit governance inputs such as trust and strategic weights. It is executed through the Strategic Data Sourcing Agent Ecosystem (SDSAE), which bypasses costly model retraining for assets that fail governance or redundancy checks. In a credit‑card fraud benchmark, only the genuinely differentiated asset passed the significance gate and received an acquisition recommendation, while a fully redundant asset scored lowest.
By Sukesh Kumar, Dilip Singh
The paper demonstrates that any Transformer block—comprising attention, LayerNorm, and a GELU MLP—can be exactly compiled into a generic deep neural network using only fixed affine maps, routing, and a small set of nonlinear functions (square, sigmoid, inverse square root, and GELU or its sigmoid approximation). It shows that removing activation‑activation products incurs an $N^2$ cost per head for dense scores, but this can be avoided by using element‑wise operations available on most NPUs. Additionally, the authors prove that exact or conditional‑mean tail‑mass estimation provides an unbiased, low‑error approximation of top‑$k$ attention, and that clustering queries or agents into centroids reduces the number of computed scores from $N^2$ to $Nk$ per head, achieving comparable performance to reference models on DeiT‑Tiny, MambaIRv2‑Light, and Pythia‑70M.
By Sergey Salishev, Zeev Volkovich, Oleg Granichin
The paper introduces the reflexivity matrix Ω to capture how one agent’s deployment influences another’s best response in multi‑agent performative prediction. It identifies a spectral threshold—when the spectral radius of Ω exceeds one—separating stable learning from worst‑case learning obstruction. Below this threshold, repeated retraining converges geometrically to a unique stable equilibrium with bounded excess loss; above it, cumulative excess loss can grow linearly with the horizon.
By Ahmed Mahrous, Roberto Di Pietro
The paper introduces DriveMLP, a lightweight MLP policy for autonomous driving that is trained using an Expert‑Guided RL (EGRL) framework. EGRL leverages reinforcement learning solely as an optimization tool, guided by a trajectory‑distance reward derived from logged expert data, and the study shows that the training strategy itself yields significant performance gains over model complexity. DriveMLP achieves an 86.51 score on the nuPlan Val14 Reactive benchmark, surpassing the previous best IL planner by 3.2 points while offering more than 15× faster inference.
By Alessandro Canevaro, Julian Schmidt, Hang Yu, Ye Jin, Sajad Marvi, Georg Martius, Julian Jordan
CORD is a conflict‑aware optimization framework that addresses gradient cancellation between task rewards and teacher guidance in joint reinforcement learning and self‑distillation. It separates the protected objective from the projection trigger, applies a nearest‑point closed‑form update per mini‑batch, and offers gradient diagnostics to identify and correct conflicts. Experiments across three training recipes show that CORD improves ALFWorld success by 1.5–15.9 percentage points and raises WebShop SDAR success from 60.7% to 64.6%.
By Wenhao Yu, Zhengcheng Wang, Jiahong Liu, Zexuan Qiu, Minda Hu, Shaohang Wei, Jiafei Lyu, Junyan Lyu, Haobo Fu, Irwin King
The paper investigates a new vulnerability called memory injection, where malicious prompts can persist across sessions in modern LLM‑as‑a‑service platforms. By demonstrating that all NIST attack goals can be achieved on ChatGPT Instant 5.3, Claude Sonnet 4.6, and Gemini 3 Flash, the authors show that these services can spread disinformation, generate harmful content, or cause offline unresponsiveness even in fresh sessions. The study also reveals that many proposed defenses are not yet implemented and provides datasets and artifacts for further research.
By Pedro H. Barcha Correia, Matheus Azevedo de S\'a, Paulo H. Gomes de Senna, Giovanni Mandel Martignago, Victor Takashi Hayashi, Charles Christian Miers, Marcos A. Simplicio Jr
The paper introduces a coding agent that selects and expands code edits before evaluation by scoring proposed changes using a linear readout of the parent’s performance and the edit’s features. It compares several selection strategies—including Huxley‑Gödel Machine selection, parent‑rate ablation, and calibration‑best edits—across HumanEval+ and Polyglot benchmarks, reporting held‑out performance for each. Experiments with GPT‑4o and Claude Sonnet 4.5 show that the proposed methods achieve high accuracy, with the best nodes reaching 87.0% on HumanEval+.
By Erfan Darzi, Sree Bhargavi Balija, Sheng He
The paper introduces ComRSI, a framework that recursively refines a semantic communication program using large‑language‑model‑assisted code edits. By iteratively modifying the encoder, channel adapters, decoder, loss, and training policy, the method achieves significant performance gains—e.g., 74.3% test accuracy on CIFAR‑100 over an AWGN channel with only 32 symbols per image, compared to 67.2% with 256 symbols in the baseline. Experiments across ten generations and repeated searches demonstrate that recursive optimization outperforms non‑recursive agents, though the advantage over a history‑equipped agent remains statistically inconclusive.
By Yan Gong, Zheng Chu, Wei Koong Chai, Hari Mohan Pandey, Jingrui Hou, Miao Zhang, Yuanjian Li, Pei Xiao
HGM-Dual is a diagnostic‑guided framework that improves harness optimization for planning agents by representing agent evolution as lineage‑aware context and using error‑conditioned dual‑path child generation. It records ancestor changes, behavior outcomes, and ineffective modifications, enabling explicit edit‑outcome attribution. The method achieves up to 14.45% relative improvement in case accuracy while using 2×–4× fewer evaluation budgets compared to the strongest baseline.
By Sudipta Paul, Pengrui Quan, Aounon Kumar, Nicholas Tzou, Vivek Kulkarni, Srinivas Chappidi
The paper introduces an energy-based transfer learning approach for reinforcement learning, aiming to improve sample efficiency in continual learning scenarios. By using out-of-distribution detection, the method selectively provides teacher guidance only in states that fall within the teacher’s training distribution, preventing sub-optimal bias when tasks differ significantly. The authors theoretically connect energy scores to state-visitation patterns under on-policy RL and demonstrate empirical gains in both single-task and multi-task settings.
By Zeyun Deng, Jasorsi Ghosh, Fiona Xie, Kai Cheng, Yuzhe Lu, Katia Sycara, Joseph Campbell
RelAgent is an LLM-based autonomous data scientist designed for relational learning. It operates in two phases: first, the LLM agent constructs SQL feature programs and selects a predictive model using database, validation, and evaluation tools; second, the resulting program is executed without further LLM calls. The final predictor combines SQL queries and a classical model, providing fast, deterministic, and interpretable predictions that rely solely on human-readable query-defined features.
By Xingyue Huang, Louis Tichelman, Jinwoo Kim, Krzysztof Olejniczak, \.Ismail \.Ilkan Ceylan
The paper demonstrates that a minimal foraging swarm can approach the explore/exploit optimum of reinforcement learning without any learner or reward signal. Agents use heavy‑tailed Levy flights to explore, recruit neighbors via a threshold cascade on their proximity network, and disperse as patch crowds, creating a cycle that balances exploration and exploitation. The swarm attains the shape of the ideal‑free distribution (committed foragers track richness with slope 0.94) and reaches this allocation dynamically in about 150 steps, though it remains slightly exploitable and does not achieve the exact equilibrium.
By Galen J. Wilkerson
DynaWeave is an automated framework that converts static tasks into dynamic scenarios by generating exogenous events and using DAGs to manage dependencies. It activates these events during agent interaction and employs a dynamic verifier that updates only relevant evaluation checks. Using DynaWeave, the authors collected over 2,000 verified trajectories, fine‑tuned Qwen3.5 models, and achieved significant performance gains on Gaia2, CostBench, and ToolMaze, even surpassing Gemini‑3.1‑Pro‑Preview on ToolMaze.
By Xuan Yang, Chunxiao Liu, Hongsheng Xin, Hao Xu, Ning Miao
The paper introduces BiasGuard‑D, a system that uses two adversarial language‑model agents from different families to exchange typed claim documents, with a third model quorum voting to adjudicate. A combined referential‑plus‑semantic invariant validates the verdict, and the iterative deliberation layer shifts remediation from stereotypes to hallucinations, improving right‑answer recovery by 49 percentage points on HaluEval QA. All claims, votes, and instructions are stored as typed records, providing an auditable provenance trail for operators under emerging AI regulations.
By Kardelen Subasi, Jaros{\l}aw A. Chudziak
PRISM is a training‑free framework for financial information retrieval that combines prompt engineering, in‑context learning, and a candidate refinement module, with optional multi‑agent coordination. On FiQA‑2018 it improves Recall@100 from 0.7783 to 0.8326, and on FinAgentBench it achieves a Cohere‑refined top‑50 pool that raises NDCG@5 while cutting per‑query cost by about 35%. The authors also release an expert‑annotated evaluation set, a label‑robustness sweep, and a Pareto frontier of quality‑cost‑latency trade‑offs, and PRISM ranks third on the private FinAgentBench leaderboard without fine‑tuning.
By Chun Chet Ng, Jia Yu Lim, Zhen Hao Chu, Yixi Zhou, Wei Zeng Low
arXiv:2610. 29845v1 Announce Type: cross Abstract: Autonomous mobile robots operating in dense crowds and cluttered sites require $360^\circ$ perception to react to surrounding hazards.
By Samuel Foo Enze, Manthan Patel, Fan Yang, Linus Kramer, Pascal Roth, Marco Hutter, Cesar Cadena