EviSeek is a training‑free framework that explicitly links question‑derived evidence demands to observation‑conditioned actions and selective evidence retention. By making the informational purpose of interaction explicit, it guides agents to acquire only the evidence needed to answer a question. On the ESI‑Bench dataset, EviSeek improves accuracy by 7.1 percentage points over a single‑model agent with Qwen3‑VL‑32B and by 3.7 points with Gemini 3.1 Pro, while also boosting visual‑support‑weighted accuracy and reducing token usage.
arXiv:2509. 23071v2 Announce Type: replace-cross Abstract: Retrieval-augmented generation (RAG) agent development is hindered by the lack of executable ground-truth agent-environment interaction trajectories.
JRDB-AVR is a new benchmark for embodied agents that tests active visual reasoning in real-world environments. It requires agents to answer questions by requesting limited observations in time and view, and evaluates both the final answer and the visual evidence that supports it. The benchmark includes diverse tasks such as temporal search, viewpoint selection, and human-oriented compositional reasoning, and introduces a reference agent that builds an observation‑grounded graph world model.
Iron is a new framework for training generalist virtual agents that aligns low‑level actions with high‑level intents using a stepwise cycle‑consistent reward. It also repurposes failed trajectories through a hindsight reproduction mechanism to improve learning efficiency and task diversity. Experiments show Iron‑trained agents outperform those trained with three times more data, achieving a 25.06% relative improvement on unseen web tasks and better performance on complex tasks.
arXiv:2606. 14772v1 Announce Type: cross Abstract: Aerial Embodied Question Answering (EQA) requires Unmanned Aerial Vehicles (UAVs) to actively perceive the environment and answer natural language questions.
arXiv:2607. 13621v1 Announce Type: new Abstract: Language-guided human following is an important capability for embodied agents, but existing benchmarks typically assume that the target person is visible at the start of an episode.
arXiv:2606. 14778v1 Announce Type: cross Abstract: Long-term action anticipation (LTA) aims to predict an ordered sequence of future verb-noun actions from a partially observed video.
The paper introduces FoG, a framework that trains knowledge graph (KG) agents to retrieve the evidence behind their answers. FoG combines evidence‑grounded synthesis, a KG interaction agent, and an evidence‑coverage process reward (ECPR) to ensure that every answer is supported by retrieved triples. Experiments on three KGQA benchmarks show that 3B–4B open‑source agents trained with FoG outperform larger baselines in both answer accuracy and evidence coverage, even with zero‑shot cross‑KG transfer.
arXiv:2606. 16432v1 Announce Type: cross Abstract: User instructions are often underspecified because humans rely on implicit assumptions about the surrounding environment.
arXiv:2608. 06128v1 Announce Type: new Abstract: Search agents extend large language models beyond static parametric memory by enabling them to acquire and use ex ternal evidence during multi-step reasoning.
arXiv:2608.22971v2 Announce Type: replace Abstract: Embodied AI has emerged as a key paradigm for autonomous systems operating in complex physical environments. While end-to-end models have establish...
arXiv:2601.16400v2 Announce Type: replace Abstract: Real-world visual question answering (VQA) may be context-dependent: an image-question pair could be under-specified, such that the correct answer...
arXiv:2609.37349v1 Announce Type: cross Abstract: Multi-step visual retrieval-augmented generation (RAG) answers complex questions by repeatedly retrieving visual evidence, updating an intermediate s...