Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2602.08370v2 Announce Type: replace-cross Abstract: Realizing versatile and human-like performance in high-demand sports like badminton remains a formidable challenge for humanoid robotics. Unl...
CHAI for LLMs is a framework that improves large language models’ performance on code‑mixed translation tasks by using LLMs as annotators to create preference data, applying reinforcement learning from AI feedback, incorporating LLM‑generated domain knowledge for iterative refinement, and evaluating on real‑world datasets. The approach yields a 68.45% average win rate over state‑of‑the‑art open‑source models in human‑adjudicated tests. It demonstrates a scalable method to enhance code‑mixed language understanding in open‑source LLMs.
arXiv:2508.02312v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs), now a foundation in advancing natural language processing, power applications such as text generation, machine...
Zonal RL-RRT is a new path‑planning algorithm that partitions a map into zones using kd‑tree partitioning and employs Value Iteration as a high‑level decision maker. The method achieves a three‑fold improvement in time efficiency over basic sampling methods such as RRT and RRT* in forest‑like maps, and outperforms heuristic‑guided methods like BIT* and Informed RRT* by 1.5× in runtime while maintaining robust success rates across 2D to 6D environments. It also shows on average a 1.5× better performance than learning‑based methods such as NeuralRRT* and MPNetSMP, and has been validated in simulations of a UR10e arm manipulator in MuJoCo.
arXiv:2609.15131v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) require substantial computation to process numerous visual tokens across all transformer layers. Most method...
arXiv:2609.15361v1 Announce Type: cross Abstract: Effective communication is a cornerstone of distributed intelligence in Multi-Agent Reinforcement Learning (MARL), yet ensuring that generated messag...
The paper discusses the agent-centric general value function (ACGVF) framework, which allows an agent to decide both which goal to pursue and when to consider a goal finished, beyond merely selecting actions. It notes that ACGVF assumes full observability, while a prior approach used an internal belief state but required externally supplied goals. The note proposes to unify and extend these methods using hierarchical hidden Markov models (HHMMs).
The study presents a virtual imaging trial framework that uses reinforcement learning to optimize computed tomography (CT) protocols, balancing liver lesion detectability against radiation dose. By training a Proximal Policy Optimization agent on 63 computational human models across 468 parameter combinations, the authors demonstrate that evaluating only eight protocols per patient—about 2% of exhaustive testing—recovers 98.2% of the optimal objective. Conditioning the agent on patient‑specific CT localizer embeddings further improves zero‑simulation recovery by 10.7 percentage points compared to a localizer‑blind policy.
The paper introduces a label‑free test‑time training approach for GUI grounding, leveraging confidence patterns in coordinate tokens rather than full‑sequence confidence. It proposes Confidence‑Anchored Learning (CAL) to filter pseudo‑labels and assign distance‑based binary rewards, and extends this to Confidence‑Anchored Negative Learning (CANL) which optimizes solely on negative samples to avoid noisy positives. Experiments show that CANL‑7B achieves 92.1% on ScreenSpot‑V2 and 33.8% on ScreenSpot‑Pro, improving the base model by 8.9%.
arXiv:2609.14066v1 Announce Type: cross Abstract: Although existing multi-agent Retrieval-Augmented Generation (RAG) systems have demonstrated promise on complex multimodal reasoning tasks, they rema...
arXiv:2609.15066v1 Announce Type: cross Abstract: We present Salesforce Koa, an enterprise language model built by post-training the open-weight Nemotron-3-Super-120B foundation model with reinforcem...
arXiv:2609.14952v1 Announce Type: new Abstract: Dynamic cloud workflow scheduling must balance deadline satisfaction, container utilization, and energy consumption while dealing with stochastic task-...
arXiv:2609.14739v1 Announce Type: cross Abstract: Large language models are increasingly used in high-stakes domains such as law, where systems must ground their reasoning in retrieved evidence and a...
arXiv:2609.14647v1 Announce Type: cross Abstract: Robots, and humanoid robots in particular, are increasingly competent at individual behaviors, each obtained by training a specialized controller. A...
arXiv:2604.13175v2 Announce Type: replace-cross Abstract: Large language models can be aligned with human preferences through offline reinforcement learning (RL) on small labeled datasets. While sing...
The paper introduces a Large Language Model–enhanced Multi-Agent Reinforcement Learning framework for optimizing electric vehicle charging, station profitability, and grid stability in public charging systems. By using an LLM to select interpretable features from IoT data and dynamically balance conflicting objectives, the approach unifies grid, EV, and station optimization in a single loop. Experiments show the method outperforms existing baselines, improving market efficiency and cutting training time by more than 70%.
arXiv:2609.13739v1 Announce Type: cross Abstract: Language-model agents are increasingly deployed through diverse harnesses that differ in system prompts, tool schemas, control loops, and trajectory...
arXiv:2609.13545v1 Announce Type: cross Abstract: Learning-based adaptive control of robotic manipulators with non-observable friction memory has been addressed by attention- based meta-controllers w...
The paper demonstrates that a single example from the BBQ fairness benchmark can dramatically improve a model’s performance, raising accuracy from 79.9% to 92.9% with Group Relative Policy Optimization and to 99.0% with one-shot in-context learning. This effect is consistent across different model families and is driven by the model’s reasoning traces, which adopt a category‑agnostic "missing evidence" pattern. The authors argue that BBQ-style multiple‑choice abstention tests capture only a single structural cue and therefore do not guarantee true fairness, calling for broader evaluation suites.
CodeTS introduces a verifiable framework for generating time series from natural language by translating textual temporal descriptions into executable code, which then produces the desired series. The approach constructs aligned Text‑Code‑TS triplets for supervised initialization and employs multi‑stage execution‑based rewards to ensure code validity and time‑series quality. Experiments on eight benchmarks show that CodeTS outperforms both LLM‑based and supervised generative baselines, offering a strong zero‑shot solution for Text‑to‑TS generation.