Pith. sign in

REVIEW 2 cited by

Kefa: A Knowledge Enhanced and Fine-grained Aligned Speaker for Navigation Instruction Generation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2307.13368 v1 pith:VHKYLRAL submitted 2023-07-25 cs.CV

Kefa: A Knowledge Enhanced and Fine-grained Aligned Speaker for Navigation Instruction Generation

classification cs.CV
keywords instructionnavigationspeakergenerationkefaknowledgealignmentfine-grained
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We introduce a novel speaker model \textsc{Kefa} for navigation instruction generation. The existing speaker models in Vision-and-Language Navigation suffer from the large domain gap of vision features between different environments and insufficient temporal grounding capability. To address the challenges, we propose a Knowledge Refinement Module to enhance the feature representation with external knowledge facts, and an Adaptive Temporal Alignment method to enforce fine-grained alignment between the generated instructions and the observation sequences. Moreover, we propose a new metric SPICE-D for navigation instruction evaluation, which is aware of the correctness of direction phrases. The experimental results on R2R and UrbanWalk datasets show that the proposed KEFA speaker achieves state-of-the-art instruction generation performance for both indoor and outdoor scenes.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. AeroDuo: Aerial Duo for UAV-based Vision and Language Navigation

    cs.CV 2025-08 conditional novelty 6.0

    Two drones at different altitudes, one guided by a vision-language model and one by a local navigator, reach targets more often than single-drone baselines on a new UAV navigation benchmark.

  2. GoViG: Goal-Conditioned Visual Navigation Instruction Generation via Multimodal Reasoning

    cs.CV 2025-08 unverdicted novelty 6.0

    GoViG decomposes goal-conditioned navigation instruction generation into visual state prediction and instruction synthesis using an autoregressive multimodal LLM with one-pass and interleaved reasoning, showing gains ...