Nikolai Ilinykh

Hej guys, thank you for your amazing work! I just have one clarification question: is there some sort of heuristics to decide which value of greedyness one should use when...

_fc and _att features [clarification question]

5

Hej, thank you for the great work! Much appreciated! I have a question about `cocobu_fc` / `cocobu_att` and `cocotalk_fc` / `cocotalk_att` pairs: why did you specifically extract fc features here?...

KneserNeyInterpolated has problem with OOV words during testing and perplexity is always inf

7

nltk 3.8.1 I am training and testing a language model on my corpus of sentences using KneserNeyInterpolated. It looks like the nltk implementation of this smoothing algorithm does not know...

Key 'defrost' not in 'HabitatConfig' when evaluating EQA-CNN pre-trained model

3

Hello, I am trying to evaluate the pre-trained EQA-CNN model on the EQA dataset. > python -u -m habitat_baselines.run --config-name=eqa/il_eqa_cnn_pretrain.yaml habitat_baselines.evaluate=True I am getting the following error: > omegaconf.errors.ConfigAttributeError: Key...

Hello, The Google Drive link to get the pre-trained CNN model for the EQA task is not working: https://github.com/facebookresearch/habitat-lab/tree/main/habitat-baselines/habitat_baselines/il#pre-trained-model Could you please update the link and/or share the model? Best,...

Pre-trained model release

1

Hello, are there any plans to release a model checkpoint used in the paper? Would be much appreciated. Thanks.

self-critical training [duration and memory occupation]

1

Hej, thank you a lot for your great work and some nice code! I have a question regarding the self-criticial extra training. I am not exactly sure if there is...

### Anthology ID 2024.clasp-1.8 ### Type of Change Revision ### PDF of the Revision or Erratum [milling-njus-updated.pdf](https://github.com/user-attachments/files/17362113/milling-njus-updated.pdf) ### Brief Description of Changes The original .pdf of the paper did not...