TechniqueRLHF / Alignment8 recent entries15 Oct 2026From Plato to Aristotle: Toward a General Theory of Representation HypothesesarXiv:2610.20286v1 Announce Type: new Abstract: The Platonic Representation Hypothesis (PRH) holds that increasingly capable models converge toward a common representation. However, recent studies hav→15 Oct 2026Discriminative Score Function: Turning Pretrained Models into Functional Generative PriorsarXiv:2610.19743v1 Announce Type: new Abstract: Discriminative models are trained with task-specific objectives such as classification, detection, or representation learning, yet their practical utili
TechniqueRAG8 recent entries9 Oct 2026Beyond Resolution: Object-to-Image Ratio Mismatch in Instance RetrievalarXiv:2610.11489v1 Announce Type: new Abstract: Visual instance retrieval often fails when the same object appears at different apparent sizes in the query and gallery. We show that the dominant cause→12 Oct 2026Knowing Is Not Telling: Identity Ossification in LLM Representations of Trans-border CommunitiesarXiv:2610.12610v1 Announce Type: cross Abstract: Ethnic and cultural identities rarely align with political borders, a tension made sharper by migration and long-standing cross-border ties. Communiti→14 Oct 2026ECHO: Correcting Time Series Foundation Model Forecasts with Historical Analogues and ErrorsarXiv:2610.16433v1 Announce Type: new Abstract: Time Series Foundation Models (TSFMs) provide general-purpose forecasting capabilities, yet their forecasts can remain misaligned with the temporal char→14 Oct 2026Cite Pretrain: Retrieval-Free Knowledge Attribution for Large Language ModelsarXiv:2506.17585v4 Announce Type: replace-cross Abstract: Large language models can answer questions correctly and still hallucinate their citations. Most citation systems address this by retrieving d→15 Oct 2026Enabling Validation for Robust Few-Shot RecognitionarXiv:2506.04713v4 Announce Type: replace Abstract: A key challenge in Few-Shot Recognition (FSR) is the scarcity of labeled task-specific data for training. Contemporary methods mitigate this challen→15 Oct 2026Dataset-Level Regularization for Joint-Embedding Self-Supervised LearningarXiv:2610.20158v1 Announce Type: new Abstract: Self-supervised joint embedding learning aligns representations of augmented views while regularizing their distribution to prevent collapse. Regulariza→15 Oct 2026Co-adaptive Rule Routing and Constraining for Structured Knowledge RetrievalarXiv:2610.14465v1 Announce Type: cross Abstract: Structured knowledge retrieval acquires task-relevant evidence from structured knowledge sources for downstream tasks. Existing methods rely on uncons→15 Oct 2026Can LLMs Access their World Knowledge for Event Prediction?arXiv:2610.19021v1 Announce Type: new Abstract: Recent work has questioned whether large language models (LLMs) can be useful for temporal event prediction. Despite vast pretrained world knowledge, LL
TechniqueAgents8 recent entries15 Oct 2026From Failures to Learnable Worlds: Policy-Conditioned Environment Generation for Navigation AgentsarXiv:2610.19420v1 Announce Type: cross Abstract: Simulated environments enable scalable training data generation for embodied agents, but more data does not necessarily provide what a particular agen→15 Oct 2026FACT: A Multimodal Force-Action-Track Foundation Model for Human-to-Robot DexterityarXiv:2610.19155v1 Announce Type: new Abstract: Human demonstrations offer a scalable alternative to costly robot data collection for dexterous manipulation, but most human datasets lack force measure→15 Oct 2026Cracking Mobile Manipulation: From Failure Diagnosis to Agentic Robot SystemarXiv:2610.19600v1 Announce Type: new Abstract: Recent vision-language-action (VLA) models have shown strong performance on local robot manipulation, yet their success does not readily translate to lo→15 Oct 2026Co-Evolving Acting and Forecasting for Language AgentsarXiv:2610.20597v1 Announce Type: cross Abstract: Training language agents is bottlenecked by the high cost of long-horizon rollouts, yet standard agentic reinforcement learning (RL) learns only from →15 Oct 2026Co-adaptive Rule Routing and Constraining for Structured Knowledge RetrievalarXiv:2610.14465v1 Announce Type: cross Abstract: Structured knowledge retrieval acquires task-relevant evidence from structured knowledge sources for downstream tasks. Existing methods rely on uncons→15 Oct 2026BluePixel: Toward Native 4K Agentic Image EditingarXiv:2610.19882v1 Announce Type: new Abstract: Instruction-driven image editing has reached native 4K, yet how such editors are trained remains undisclosed, and pretraining one from scratch is beyond→15 Oct 2026AdaRSI: On-Demand Recursive Self-Improvement for Embodied AgentsarXiv:2610.20474v1 Announce Type: new Abstract: Pretrained vision-language-action models (VLAs) often fail midway through unseen long-horizon tasks. Recursive self-improvement (RSI) offers a remedy, b→15 Oct 2026A Sovereign, Indic-First, Reasoning ModelarXiv:2610.15371v1 Announce Type: cross Abstract: Gnani-Evon-v3.3-30B-A3B is a long-context hybrid Mamba-Transformer mixture-of-experts model for reasoning, instruction following, and tool use in Engl
TechniqueFine-tuning8 recent entries15 Oct 2026FACT: A Multimodal Force-Action-Track Foundation Model for Human-to-Robot DexterityarXiv:2610.19155v1 Announce Type: new Abstract: Human demonstrations offer a scalable alternative to costly robot data collection for dexterous manipulation, but most human datasets lack force measure→15 Oct 2026Eye, Robot: Learning Expressive Gaze Dynamics via a VLM-Guided Transformer-VAEarXiv:2610.19635v1 Announce Type: new Abstract: Gaze coordinates attention in human interaction through coupled eye, head, and eyelid motion. A humanoid robot must choose where to look while generatin→15 Oct 2026D2: High-Capability Full-Duplex Spoken Language Models from Half-Duplex BackbonesarXiv:2610.17124v1 Announce Type: cross Abstract: Full-duplex spoken language models (FD-SLMs) support simultaneous listening and speaking, but often lag behind half-duplex spoken language models (HD-→15 Oct 2026Continual pretraining Prostate MRI ViT Encoder for task alignmentarXiv:2610.20217v1 Announce Type: new Abstract: Building foundation models for MRI has emerged as a promising direction for advancing medical image analysis and improving accessibility of task specifi→15 Oct 2026Co-adaptive Rule Routing and Constraining for Structured Knowledge RetrievalarXiv:2610.14465v1 Announce Type: cross Abstract: Structured knowledge retrieval acquires task-relevant evidence from structured knowledge sources for downstream tasks. Existing methods rely on uncons→15 Oct 2026Back-Translation Still Matters: Leveraging Monolingual Data for Low-Resource Machine Translation with LLMsarXiv:2610.18818v2 Announce Type: cross Abstract: Pre-trained Large Language Model (LLMs) are effective on many tasks in zero-shot and few-shot settings. For machine translation of low-resource langua→15 Oct 2026AdaRSI: On-Demand Recursive Self-Improvement for Embodied AgentsarXiv:2610.20474v1 Announce Type: new Abstract: Pretrained vision-language-action models (VLAs) often fail midway through unseen long-horizon tasks. Recursive self-improvement (RSI) offers a remedy, b→15 Oct 2026AcrossVAM2.0: Conflict-Aware Point Supervision for Parameter-Efficient Video Diffusion AdaptationarXiv:2610.14693v1 Announce Type: new Abstract: Explicit point correspondence is attractive for adapting video generators: it describes how visible surfaces move, but can overwrite the appearance prio
TechniqueMultimodal8 recent entries15 Oct 2026Enabling Validation for Robust Few-Shot RecognitionarXiv:2506.04713v4 Announce Type: replace Abstract: A key challenge in Few-Shot Recognition (FSR) is the scarcity of labeled task-specific data for training. Contemporary methods mitigate this challen→15 Oct 2026DynThinker: Unifying Static 3D and Dynamic 4D Spatial ReasoningarXiv:2610.20598v1 Announce Type: new Abstract: Achieving general spatial intelligence requires a unified understanding of both static 3D and dynamic 4D. However, we find that existing models struggle→15 Oct 2026Cracking Mobile Manipulation: From Failure Diagnosis to Agentic Robot SystemarXiv:2610.19600v1 Announce Type: new Abstract: Recent vision-language-action (VLA) models have shown strong performance on local robot manipulation, yet their success does not readily translate to lo→15 Oct 2026Continual pretraining Prostate MRI ViT Encoder for task alignmentarXiv:2610.20217v1 Announce Type: new Abstract: Building foundation models for MRI has emerged as a promising direction for advancing medical image analysis and improving accessibility of task specifi→15 Oct 2026Can Learning to Answer Spatial Questions Improve Novel-View Generation?arXiv:2610.20610v1 Announce Type: new Abstract: Generating novel views from limited observations requires preserving both the appearance and spatial structure of a scene. Although recent approaches ex→15 Oct 2026Beyond Size: How Gradients Shape Pruning Decisions in Large Language ModelsarXiv:2311.04902v3 Announce Type: replace Abstract: Large Language Models (LLMs) with billions of parameters are prime targets for network pruning, which removes some model weights without hurting per→15 Oct 2026Align Where the Words Look: Cross-Attention-Guided Patch Alignment with Contrastive and Transport Regularization for Bengali CaptioningarXiv:2509.18369v2 Announce Type: replace Abstract: Adapting vision-language models in low-resource languages remains challenging, as they often produce fluent text about the wrong objects. This stems→15 Oct 2026AdaRSI: On-Demand Recursive Self-Improvement for Embodied AgentsarXiv:2610.20474v1 Announce Type: new Abstract: Pretrained vision-language-action models (VLAs) often fail midway through unseen long-horizon tasks. Recursive self-improvement (RSI) offers a remedy, b
TechniqueSafety8 recent entries15 Oct 2026Human Demonstrations as Post-Training Data for Any Robot Body: Retargeting the Object, Not the HandarXiv:2610.20351v1 Announce Type: new Abstract: Human demonstrations are today used largely for pretraining vision-language-action models, with little ability to use the same data for target-embodimen→15 Oct 2026Historical-Feature Residual Adaptation for Sequential ScreeningarXiv:2610.19315v1 Announce Type: new Abstract: Sequential screening policies must use acquired feedback effectively under a fixed experiment budget. We study acquisition-time residual adaptation: a s→15 Oct 2026End-to-End Vision-Language-Action Policies for Autonomous Lunar Rover DockingarXiv:2610.19224v1 Announce Type: new Abstract: Surface navigation on the Moon is subject to extreme illumination contrast and deployment-time perceptual disturbances. While modular navigation pipelin→15 Oct 2026Continual pretraining Prostate MRI ViT Encoder for task alignmentarXiv:2610.20217v1 Announce Type: new Abstract: Building foundation models for MRI has emerged as a promising direction for advancing medical image analysis and improving accessibility of task specifi→15 Oct 2026Co-Evolving Acting and Forecasting for Language AgentsarXiv:2610.20597v1 Announce Type: cross Abstract: Training language agents is bottlenecked by the high cost of long-horizon rollouts, yet standard agentic reinforcement learning (RL) learns only from →15 Oct 2026Co-adaptive Rule Routing and Constraining for Structured Knowledge RetrievalarXiv:2610.14465v1 Announce Type: cross Abstract: Structured knowledge retrieval acquires task-relevant evidence from structured knowledge sources for downstream tasks. Existing methods rely on uncons→15 Oct 2026Asymptotically Exact Negative Guidance of Diffusion Models via Positive-Unlabeled LearningarXiv:2610.19802v1 Announce Type: new Abstract: Negative guidance of diffusion models generates samples that avoid an undesired condition, specified either by reference samples or by a conditional sco→15 Oct 2026Align Where the Words Look: Cross-Attention-Guided Patch Alignment with Contrastive and Transport Regularization for Bengali CaptioningarXiv:2509.18369v2 Announce Type: replace Abstract: Adapting vision-language models in low-resource languages remains challenging, as they often produce fluent text about the wrong objects. This stems