Qantara: Bridge-Flow Training for Multi-Paradigm JEPA Control

• URL: http://arxiv.org/abs/2607.04978v1
• Date: Mon, 06 Jul 2026 12:12:50 GMT
• Title: Qantara: Bridge-Flow Training for Multi-Paradigm JEPA Control
• Abstract summary: Joint-Embedding Predictive Architectures underpin a growing family of latent world models for control from raw pixels.<n>A single checkpoint that serves both would defer this choice to inference, when deployment constraints determine which path wins.<n>We present Qantara, an end-to-end JEPA whose joint training objective pairs a Brownian-bridge interpolant between consecutive clean latents.
• Score: 5.357451930622806
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Joint-Embedding Predictive Architectures (JEPAs) underpin a growing family of latent world models for control from raw pixels, but every existing JEPA world model commits at training time to a single inference paradigm: either trajectory optimisation in a learned dynamics model, or direct behaviour cloning. A single checkpoint that serves both would defer this choice to inference, when deployment constraints (rollout cost, observation accessibility) determine which path wins. We present Qantara, an end-to-end JEPA whose joint training objective pairs a Brownian-bridge interpolant between consecutive clean latents on the state axis with noise-to-data flow matching on the action axis. The same checkpoint serves three inference paradigms without retraining: latent planning, behaviour-cloning action sampling, and inverse dynamics, which we query through a video-inverse composition that first predicts the next latent without action conditioning, then extracts the action. Training concentrates mass on the edges of the (action-time, state-time) noise square, where inference queries the predictor: replacing it with uniform interior sampling drops Push-T planning from 90.1 to 53.3 SR at matched compute. On the LeWM control suite, Qantara reaches a 91.2 SR three-train-seed average and sets new SOTA on OGBench-Cube (+7.7 SR over DINO-WM, +19.7 over LeWM). From the same weights, the behaviour-cloning and video-inverse paths reach 82-83 SR on Push-T and 71-73 SR on Cube. These results move JEPA world models from single-paradigm planners to multi-paradigm controllers.

Related papers

• On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability [57.03393882933515]
  We describe the architecture and ablations of Qwen3.8-Flash-Next, a sparse mixture-of-experts model with 125B parameters, 6B activated per token, and 51B parameters of n-gram embedding tables held off the accelerator.<n>On fourteen pre-training benchmarks the model leads the 397B-A17B predecessor on eight and trails it on the rest by at most 2.6 points, at 1/3 the activated parameters, 1/3 the training tokens, and roughly 1/9 the training FLOPs.
  arXiv  Detail & Related papers  (2026-08-31T06:35:07Z)
• AquaJEPA: Action-Conditioned Multimodal Predictive Representations for Underwater Robot Dynamics [0.0]
  AquaJEPA fuses an RGB camera, forward-looking sonar, and proprioception with explicit sensor validity.<n>It predicts a future latent target conditioned on eight-thruster commands.<n>It significantly outperforms three action-conditioned predictive baselines in paired final error.
  arXiv  Detail & Related papers  (2026-07-31T13:11:11Z)
• QuantWAMs: Calibrating at the Right Granularity for World Action Models [34.540026172139406]
  World Action Models jointly predict future observations and actions, but their iterative denoising and closed-loop execution make deployment costly.<n>We present QuantWAMs, a PTQ framework that aligns quantization decisions with the calibration context defined by model structure, rollout distribution, and task objective.<n>We evaluate QuantWAMs on Fast-WAM and LingBot-VA across RoboTwin 2.0, LIBERO, and real-robot manipulation with an AgiBot G2.
  arXiv  Detail & Related papers  (2026-07-30T15:54:29Z)
• MoP-JEPA: Hard-Assigned Predictor Mixtures for Stochastic JEPA World Models [12.290764474660103]
  JEPA world models predict the next latent state with a single deterministic predictor trained by latent regression.<n>We show that this fails structurally when the environment is branching.<n>We prove that MoP-JEPA's hard-assigned predictors converge instead to a quantizer of the transition distribution.
  arXiv  Detail & Related papers  (2026-07-06T15:47:37Z)
• Beyond the Current Observation: Evaluating Multimodal Large Language Models in Controllable Non-Markov Games [69.57330692969543]
  We introduce RNG-Bench, a benchmark suite designed to isolate a base model's ability to reconstruct past observations.<n>RNG-Bench includes two complementary games: Matching Pairs, where card identities briefly revealed at specific locations must later be recalled, and 3D Maze, where egocentric views must be integrated into a spatial map.<n>The hardest configurations require contexts of roughly 128K tokens and 350 image inputs per episode, and remain far from saturated by frontier MLLMs.
  arXiv  Detail & Related papers  (2026-06-17T17:59:34Z)
• D$^3$-MoE:Dual Disentangled Diffusion Mixture-of-Experts for Style-Controllable End-to-End Autonomous Driving [22.70855312181222]
  D$3$-MoE disentangles trajectory modeling along two complementary axes.<n>On the behavioral axis, generation is decoupled from selection.<n>On the physical axis, decoupled longitudinal and lateral routers activate their respective experts.
  arXiv  Detail & Related papers  (2026-06-03T13:46:46Z)
• DriveAnchor: Progressive Anchor-based Flow Learning for Autonomous Driving Planning [6.848508447210393]
  DriveAnchor is a framework for autonomous driving planning.<n>It achieves behavioral diversity, controllability, and safety in a composable pipeline.<n>It reduces near-range collision rates by 89% and improves mean reward by 32% without degradation in imitation accuracy.
  arXiv  Detail & Related papers  (2026-05-30T04:17:59Z)
• Subspace-Decomposed JEPAs: Disentangling Progression and Content in Latent World Models [1.034052616244602]
  Joint-Embedding Predictive Architectures learn compact latent world models by predicting future embeddings.<n>We carve the JEPA latent into two subspaces with disjoint roles.<n>We prove that the two anti-collapse forces act on disjoint coordinates, so they compose additively rather than competing on the same dimensions.
  arXiv  Detail & Related papers  (2026-05-29T10:23:48Z)
• UWM-JEPA: Predictive World Models That Imagine in Belief Space [0.2864713389096699]
  We introduce the Unitary World Model JEPA, a JEPA world model with a density-matrix latent on a joint system-environment space and a learned unitary predictor.<n>The construction preserves the joint-state spectrum exactly during rollout, so the predictor itself cannot dissipate the represented uncertainty.<n>For JEPA world models to imagine under partial observability, latent geometry and predictor dynamics matter, not frozen context-encoding capacity alone.
  arXiv  Detail & Related papers  (2026-05-25T00:28:51Z)
• Unextractable Protocol Models: Collaborative Training and Inference without Weight Materialization [58.14514930760722]
  We consider a decentralized setup in which the participants collaboratively train and serve a large neural network.<n>In this setup, we explore the possibility of unmaterializable weights, where a full weight set is never available to any one participant.<n>We introduce Unextractable Protocol Models (UPMs): a training and inference framework that leverages the sharded model setup.
  arXiv  Detail & Related papers  (2026-05-22T10:24:57Z)
• Entropy Across the Bridge: Conditional-Marginal Discretization for Flow and Schrödinger Samplers [68.9946633641494]
  Flow matching and Schrdinger bridges define probability, yet their inference grids are usually or inherited from one-endpoint diffusion.<n>We derive a conditional-marginal entropy-rate objective for bridge-aware discretization, separating endpoint-conditioned bridge geometry from marginal flow evolution.<n>On EDM/CIFAR-10, the entropic time-discretization gives the best tested five-step FID (186.3 pm 4.0 versus 200.5 pm 2.9 for linear and 238.0 pm 5.3 for cosine)
  arXiv  Detail & Related papers  (2026-05-15T16:11:10Z)
• SOAR: Self-Correction for Optimal Alignment and Refinement in Diffusion Models [48.335262141752715]
  Post-training pipeline for diffusion models has two stages: supervised fine-tuning (SFT) on curated data and reinforcement learning (RL) with reward models.<n>We propose SOAR (Self-Correction for Optimal Alignment and Refinement), a bias-correction post-training method that fills this gap.<n>It is on-policy, reward-free, and provides dense per-time supervision with no credit-assignment problem.
  arXiv  Detail & Related papers  (2026-04-14T11:45:15Z)
• LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels [49.35636088613484]
  Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces.<n>We introduce LeModelWorld, the first JEPA that trains stably end-to-end from raw pixels using only two loss terms.<n>With 15M parameters trainable on a single GPU in a few hours, LeWM plans up to 48x faster than foundation-model-based world models.
  arXiv  Detail & Related papers  (2026-03-13T19:48:14Z)
• Agentic World Modeling for 6G: Near-Real-Time Generative State-Space Reasoning [70.56067503630486]
  We argue that sixth-generation (6G) intelligence is not fluent token prediction but calibrated the capacity to imagine and choose.<n>We show that WM-MS3M cuts mean absolute error (MAE) by 1.69% versus MS3M with 32% fewer parameters and similar latency, and achieves 35-80% lower root mean squared error (RMSE) than attention/hybrid baselines with 2.3-4.1x faster inference.
  arXiv  Detail & Related papers  (2025-11-04T17:22:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.