Search results

how to process negative image data for object detection or grounding task?

Status: Open.

#74 In NVlabs/Eagle;

· zzl777 opened on Jun 11, 2026

LM Studio

Status: Open.

#73 In NVlabs/Eagle;

· ayuso2013 opened on Jun 9, 2026

LM Studio

Status: Open.

#72 In NVlabs/Eagle;

· ayuso2013 opened on Jun 9, 2026

NotImplementedError: self._attn_implementation='flash_attention_2'

Status: Open.

#71 In NVlabs/Eagle;

· pvtoan opened on Jun 9, 2026

locate anything backbone generalization

Status: Open.

#70 In NVlabs/Eagle;

· 36330 opened on Jun 9, 2026

(LocateAnything) Label corruption and incorrect detections observed in 18-class object detection

Status: Open.

#69 In NVlabs/Eagle;

· 0xNOY opened on Jun 6, 2026

OutOfMemoryError in single L40 GPU

Status: Open.

#68 In NVlabs/Eagle;

· dianjingshilezhi opened on Jun 4, 2026

The effect is not as good as Rexonni

Status: Open.

#67 In NVlabs/Eagle;

· ffxxy opened on Jun 4, 2026

Question about dataset annotation workflow

Status: Open.

#66 In NVlabs/Eagle;

· jack-zhen opened on Jun 4, 2026

out of memory

Status: Open.

#65 In NVlabs/Eagle;

· Simon-liusheng opened on Jun 3, 2026

Discovered during self-quantization: The model is mistakenly registered as Qwen3, but it is actually based on Qwen2.5

Status: Open.

#64 In NVlabs/Eagle;

· dayuan12138 opened on Jun 2, 2026

example.jpg is missing, in locateanything_work.py

Status: Open.

#60 In NVlabs/Eagle;

· quantbruce opened on May 30, 2026