fix missing links

Author: garrett4wade

docs/tutorial/gsm8k_grpo.md

@@ -453,9 +453,7 @@ GPU 3: SGLang        GPU 7: FSDP rank 3  ─┘
                      NCCL AllReduce for gradients
 ```
 
-Each worker processes its shard, then synchronizes gradients via NCCL. For custom
-algorithms, see
-[Customization: Algorithms](https://inclusionai.github.io/AReaL/customization/algorithm.html).
+Each worker processes its shard, then synchronizes gradients via NCCL.
 
 ### The Training Loop
 
@@ -599,12 +597,11 @@ Now that you understand the basics, explore these advanced topics:
 - [Evaluation](../tutorial/eval.md) - Evaluate your trained model
 - [Training Large MoE Models](../tutorial/megatron.md) - Scale to massive models with
   Megatron integration
-- [Agentic RL with OpenAI APIs](../tutorial/agentic_rl.md) - Build agents that use tools
-  and APIs
+- [Agentic RL](../tutorial/agentic_rl.md) - Build agents that use tools and any agentic
+  frameworks
 
 **Customization Guides**:
 
 - [Custom Datasets](../customization/dataset.md) - Use your own data sources
 - [Custom Workflows](../customization/agent.md) - Build agentic/RLVR workflows with
   custom reward functions
-- [Custom Algorithms](../customization/algorithm.md) - Implement your own RL algorithms

docs/tutorial/quickstart.md

@@ -155,4 +155,3 @@ Customization guides:
 
 - [Custom dataset](../customization/dataset.md)
 - [Custom agentic/RVLR rollout workflows](../customization/agent.md)
-- [Custom algorithms](../customization/algorithm.md)

examples/tau2/train.py

@@ -9,8 +9,8 @@
 
 from examples.tau2.utils import Tau2PPOConfig
 
+from areal import PPOTrainer
 from areal.api.cli_args import load_expr_config
-from areal.experimental.trainer.rl import PPOTrainer
 from areal.utils import logging
 
 logger = logging.getLogger("Tau2Train")