adposition_ids argument for verl

modeling_hyperclovax.py

@@ -520,6 +520,7 @@ class HCXVisionForCausalLM(PreTrainedModel, GenerationMixin):
         pixel_values_videos: Optional[List[List[torch.FloatTensor]]] = None,
         past_key_values: Optional[Tuple[Tuple[torch.Tensor]]] = None,
         attention_mask: Optional[torch.FloatTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
         inputs_embeds: Optional[torch.FloatTensor] = None,
         labels: Optional[torch.LongTensor] = None,
         use_cache: Optional[bool] = None,
@@ -596,6 +597,7 @@ class HCXVisionForCausalLM(PreTrainedModel, GenerationMixin):
             input_ids=input_ids,
             inputs_embeds=inputs_embeds,
             attention_mask=attention_mask,
+            position_ids=position_ids,
             past_key_values=past_key_values,
             use_cache=use_cache,
             output_attentions=output_attentions,