shikiw/OPERA

Issues

• Request for Release of `GPT-4 assisted evaluation` Code

  #45 opened 23 days ago by SDaoer
  1

• 热力图相关

  #44 opened 24 days ago by sunzjz
  1

• 请问是否可以在LMDeploy、vLLM等部署框架上使用？

  #43 opened a month ago by Shubao-Zhao
  0

• Issue about the visual case provided in the paper

  #41 opened a month ago by Adherer
  1

• 您好，请问utils.py里面以下代码都为None，是什么导致的

  #42 opened a month ago by zhly0
  6

• Issue about visualization

  #40 opened a month ago by Melon-Xu
  6

• Discrepancy in Random Split Numbers for POPE (#2910 vs #3000)

  #38 opened a month ago by mouxingyang
  1

• Can not reproduce the results on LLaVA-1.5

  #39 opened a month ago by wangyouze
  3

• CUDA error: out of memory

  #36 opened 2 months ago by Melon-Xu
  2

• Random 500 samples in MSCOCO

  #37 opened 2 months ago by Adherer
  2

• Random 500 samples in MSCOCO?

  #12 opened 5 months ago by zhaoshitian
  4

• 可视化问题

  #34 opened 2 months ago by OliverLeeXZ
  4

• Questions about function prepare_inputs_labels_for_multimodal

  #24 opened 2 months ago by KlaineWei
  1

• CHAIR hallucination evaluation

  #30 opened 2 months ago by running-alpaca
  2

• 关于可视化

  #32 opened 2 months ago by yubo97
  2

• 关于复现POPE的结果问题

  #33 opened 2 months ago by 1906200111
  14

• CHAIR Reproduction Bugs

  #20 opened 5 months ago by xing0047
  6

• GPU information

  #21 opened 5 months ago by KlaineWei
  4

• 关于是否是幻觉句子的问题

  #31 opened 2 months ago by clclclaiggg
  2

• 是否能支持4.37.2 的transformers

  #16 opened 5 months ago by awzhgw
  2

• Does the method to find `knowledge aggregation pattern` have any relevant papers to reference in NLP domain?

  #26 opened 3 months ago by shanpoyang654
  1

• Truncation of generated results

  #27 opened 3 months ago by lalulxm
  3

• Could you provide the 'model.generate_output' function?

  #28 opened 3 months ago by huofushuo
  2

• Shikra Version

  #29 opened 4 months ago by summoneryhl
  0

• stopping criteria

  #25 opened 4 months ago by yeonju7kim
  1

• 运行vis.ipynb报错

  #23 opened 4 months ago by Ivesfu
  1

• Questions about Figure 3 in paper

  #18 opened 4 months ago by minhoooo1
  1

• Over-Trust Logit Penalty

  #17 opened 4 months ago by hubujy
  1

• AssertionError: OPERA does not support beam=1 in the current version. It will be added in the future.

  #15 opened 4 months ago by FanshuoZeng
  1

• Could you provide the script to plot the attention map?

  #14 opened 5 months ago by Rainlt
  9

• reproducing shikra problem

  #22 opened 5 months ago by KlaineWei
  0

• reproducing the result

  #19 opened 5 months ago by JEONG8652
  2

• Support for interleaved image-text comprehension(multi-image)

  #13 opened 5 months ago by laserwave
  1

• can you add support for other models like QWEN-VL in the future?

  #9 opened 5 months ago by chuangzhidan
  10

• Attention map plotting

  #10 opened 5 months ago by franciscoliu
  1

• Acknowledge LAVIS

  #11 opened 5 months ago by dxli94
  1

• What should key_position be on mPLUG-Owl2?

  #7 opened 5 months ago by BillChan226
  1

• Does it work well on videos?

  #1 opened 5 months ago by YajieW99
  1

• Reproducing MiniGPT-4's POPE result

  #8 opened 6 months ago by Ocean-627
  4

• opera_greedy_search implemented ?

  #6 opened 7 months ago by BillChan226
  1

• Model inference speed is slow

  #3 opened 7 months ago by hlz0606
  2

• AttributeError: 'MiniGPT4' object has no attribute 'embed_tokens'

  #5 opened 9 months ago by BillChan226
  2

• 'MiniGPT4' object has no attribute 'embed_tokens'

  #4 opened 9 months ago by isruihu
  0

• Questions about the IM_START and IM_END tokens

  #2 opened 9 months ago by Haotian-Zhang
  1